Scoping Study Shows Strategic Rare Earth Potential

scoping study confirms kasiya's POTENTIAL as a globally significant AND STRATEGIC source of critical rare earths

Simple, capital-efficient circuit to recover highly sought-after critical rare earths with significant valuable Nd+Pr+Dy+Tb+Y content as a potential monazite by-product - enhancing the exceptional economics of the Kasiya Rutile-Graphite DFS

Sovereign Metals Limited (ASX:SVM; AIM:SVML; OTCQX: SVMLF) (Sovereign or the Company) is pleased to announce the results of the Scoping Study (Scoping Study or Study) for recovery of a monazite concentrate with significantly elevated heavy rare earths content as a by-product from its Kasiya Project in Malawi. Concurrent with the Study, the Project has been renamed the Kasiya Critical Minerals Project (Kasiya or the Project), reflecting its combined rutile (titanium), graphite, and rare earth product suite.

CAUTIONARY STATEMENTS The Scoping Study referred to in this announcement assesses the potential incremental recovery of monazite from mineral streams that potentially may be generated by the Kasiya Project as included in the April 2026 Definitive Feasibility Study ( DFS ). The Scoping Study is based on low-level technical and economic assessments, supported by project-specific metallurgical testwork and the existing DFS production schedule. Consistent with the preliminary nature of a Scoping Study, it is not sufficient to support the estimation of Ore Reserves for the additional products or to provide assurance of an economic development case for those products at this stage. Further evaluation and appropriate studies are required before the additional products can be assessed as to whether they can be estimated as Ore Reserves and be incorporated into a definitive development case. The Scoping Study has been prepared to an accuracy level of ±30%. The results should not be considered a profit forecast or production forecast. The production profile for the production target and forecast financial information of this Study is supported by the existing April 2026 DFS production schedule (excluding years with >50% of plant feed in Inferred category of Mineral Resources). Approximately 99% of the scheduled throughput (production target) of this Scoping Study for the first twelve and half years of estimated production is in the Indicated Mineral Resource category with only 1% in the Inferred Mineral Resource category. For the life of mine of this Scoping Study, approximately 72% of life of mine production target is in the Indicated Mineral Resource category and 28% is in the Inferred Mineral Resource category. The Company has concluded that it has reasonable grounds for disclosing a production target which includes a modest amount of Inferred material and over the life of the mine the amount of Inferred Resources is not the determining factor in project viability. However, there is a low level of geological confidence associated with Inferred mineral resources and there is no certainty that further exploration work on the Kasiya deposit will result in the determination of additional Indicated Mineral Resources or that the production target (or financial forecasts) itself will be realised. The Scoping Study is based on the material assumptions outlined elsewhere in this announcement. These include assumptions about the availability of funding. It also relies on Scoping Study-level assumptions regarding metallurgical recoveries, product yields and quality, costs, prices, payability and logistics. The assessment is incremental to, and integrated with, the underlying Kasiya DFS project and therefore assumes that the DFS project is funded, developed and operated broadly as scheduled (refer to announcement dated 16 April 2026). While the Company considers the material assumptions to be based on reasonable grounds, there is no certainty that they will prove to be correct or that the range of outcomes indicated by the Scoping Study will be achieved. As a Scoping Study, the estimates and outcomes remain subject to the accuracy range, qualifications and risks disclosed in this announcement and to refinement through further technical and commercial work. To achieve the range outcomes indicated in the Scoping Study, additional funding will likely be required. Investors should note that there is no certainty that Sovereign will be able to raise funding when needed. It is also possible that such funding may only be available on terms that dilute or otherwise affect the value of the Sovereign's existing shares. It is also possible that Sovereign could pursue other 'value realisation' strategies such as sale, partial sale, or joint venture of the Project. If it does, this could materially reduce Sovereign's proportionate ownership of the Project. The Company has concluded it has a reasonable basis for providing the forward looking statements included in this announcement and believes that it has a reasonable basis to expect it will be able to fund the development of the Project. Given the uncertainties involved, investors should not make any investment decisions based solely on the results of the Scoping Study.

Accuracy range above of ±30%

Managing Director and CEO, Mr Frank Eagar, commented:

"The Kasiya DFS already defines one of the world's largest natural rutile and flake graphite developments. This Study shows the same resource can potentially deliver rare earths and deliver them cheaply. For ~US$29 million of initial capital, using infrastructure the DFS has already designed and costed, we can potentially add ~US$722 million of pre-tax value at a ~151% rate of return, with payback in around 18 months.

Mineral-reliant industries in the United States are worth over US$4 trillion a year - more than an eighth of the entire U.S. economy - and rare earths sit at the centre of them. Last year the U.S. doubled its rare earth production to almost 9,000 tonnes - and still imported 67% of what it consumed. For heavy rare earths like dysprosium, terbium and yttrium, the U.S. produces none at commercial scale and remains completely dependent on imports. That is why Washington is putting stockpiles, price floors and offtake vehicles in place.

Kasiya can potentially provide the supply this effort requires: five of the seven rare earths under Chinese export controls, for at least 23 years, from a project already at the definitive feasibility stage. And because rutile and graphite carry the cost base, our rare earths do not need high rare earth prices or floor prices to be viable.

We will now take these results into our discussions with potential partners and progress the integration of rare earths into Kasiya's definitive development case."

HIGHLIGHTS

Kasiya - A Rare Earths Project With No Mine Of Its Own To Build

o 2,626tpa of monazite rare earth concentrate (REC) containing 1,485tpa of total rare earth oxides (TREO) at steady state run of mine

Commercially Significant Levels Of China's Export-Controlled Rare Earths

o Scarce dysprosium (Dy), terbium (Tb), yttrium (Y), samarium (Sm) and gadolinium (Gd) oxides in elevated levels - all under Chinese export controls since April 2025

o Neodymium-praseodymium (NdPr), the primary input to Neodymium-Iron-Boron (NdFeB) permanent magnets, also present in levels similar to other Western rare earths mines

o 310tpa of NdPr and 36tpa DyTb - sufficient for the magnets in ~7 million humanoid robots over LOM; DyTb alone equivalent to ~18% of nameplate feed of America's first DyTb separation plant

o 193tpa Y - military radar and jet-engine thermal-barrier coatings; equivalent to ~35% of average annual U.S. consumption, which is 100% import-reliant

o 82tpa Sm + Gd - samarium used to steer Patriot, Tomahawk and AMRAAM missiles depleted to 2-3-year supply; gadolinium equivalent to 70% of the world's annual MRI supply. U.S. defence demand for both: classified

Recovered from mineral streams already generated by the DFS flowsheet

o No additional mining, no additional front-end processing, and none of the development risk of a standalone rare earths project

o Mining, processing, tailings, infrastructure, and environmental and social disciplines already assessed at DFS level with overview by Rio Tinto and Sovereign Technical Committee

operating margin provideS compelling economics

o Base Case ~US$722M Pre-Tax NPV8 uplift to Kasiya from incremental capital to first production of ~US$29M with potential upside NPV8 of US$883M

o Incremental Base Case Pre-Tax IRR of ~151% and payback of approximately 1.5 years

o ~90% operating margin with incremental site operating costs of ~US$0.90/kg REC

o ~US$84M incremental steady state annual EBITDA lifts Kasiya profitability and cash flow generation

o Pre-tax, unlevered free cash flow of ~US$1.8B over potential 23-year initial life of mine (LOM)

o Total Integrated Kasiya Pre-Tax NPV of US$2.9Bn

Maiden monazite mineral resource estimate (MRE) COMPLETED

o 524.4Mt @ 0.0132% monazite by-product underpins the Scoping Study production profile

STRONG RESULTS PROVIDE Clear path forward

o Variability testwork on TREO distribution within REC product with REC marketing and offtake discussions commencing immediately

o PFS completion in 2027

SUMMARY OF STUDY RESULTS

TABLE 1: REC SCOPING STUDY SUMMARY

• LOM in this Study differs from the 25-year DFS life of mine because Years 24 and 25 have been excluded, as they would otherwise be based solely on Inferred Resources (for which there is a low level of geological confidence), so no financial forecasts or production targets are estimated for those years and they have been excluded from the LOM for the Scoping Study. Steady state = years when run-of-mine is at nameplate capacity of 24Mtpa; all results on a 100% basis; ±30% accuracy.
• Base Case represents Argus Media group (Argus) rare earths price forecasts with payability of 50%; Western Supply Case represents Argus rare earths price forecasts and payability of 60% representing value attributable to ex-China monazite concentrate supply.

TABLE 2: SIDE-BY-SIDE KASIYA RESULTS

The addition of the rare earth circuit leaves the Kasiya DFS physicals unchanged: a 25-year initial mine life, 24Mtpa ore throughput, 222ktpa of natural rutile and 275ktpa of natural flake graphite.

SOVEREIGN ADDS RARE EARTHS TO RUTILE (TITANIUM) AND GRAPHITE PRODUCT SUITE

Sovereign has completed this Scoping Study to assess the operational, commercial and economic viability of adding a third product stream to the Kasiya Project. The Study shows that ~2,626tpa of monazite REC can be recovered from the rutile tailings stream. Together with the results of the Kasiya DFS, which was completed with input from Rio Tinto, Sovereign is potentially positioned to be a multi-decade supplier of a critical minerals products suite essential to the U.S. and U.S.-allied defence and aerospace, AI infrastructure, robotics, energy and other vital industries.

KASIYA - ONE PROJECT TO FEED THE WEST'S CRITICAL SUPPLY COMPLEX

Figure 1: Kasiya Circuit Schematic and Product Suite

Figure 2: Kasiya's life of mine production in end-use equivalents. Sources: see Appendix 4

MONAZITE BY-PRODUCT CONFIRMED WITH MAIDEN MINERAL RESOURCE ESTIMATE

Monazite mineralisation occurs within the same deeply weathered, near-surface residual heavy-mineral system that hosts the Kasiya rutile and graphite Mineral Resource as per the DFS.

To support a by-product MRE for monazite, Sovereign completed a comprehensive programme covering the April 2026 DFS mine plan: 3,250 magnetic heavy-mineral composites, representing 1,012 boreholes within the DFS pit areas, were submitted for XRF analysis, with selected composites also analysed for the full rare earth element suite by ICP to assess variations in rare earth composition across the deposit and through the weathering profile.

The maiden by-product MRE, constrained to the DFS open pits, is 69kt monazite contained in 524.4Mt at 0.0132% monazite and classified 74% Indicated and 26% Inferred. The grade reflects monazite's status as a by-product. Since the monazite MRE sits within pits the DFS has already designed, no additional mining is required to access it.

Figure 3: Monazite contained within the Kasiya 2026 MRE

PREMIUM RARE EARTHS PRODUCT CONTAINING GENUINELY RARE ELEMENTS

In January 2026, Sovereign announced that it had successfully recovered a monazite product containing high-value heavy rare earth elements (REE) alongside common light REEs from the tailings stream generated during rutile processing at its Lilongwe laboratory facilities in Malawi. The concentrate was recovered from material that would otherwise be discarded, i.e. the non-conductor tailings stream from electrostatic separation of a heavy mineral gravity concentrate of Kasiya ore. Independent preliminary chemical analysis of magnetic concentrates from processed resource drilling samples confirmed the favourable rare earth oxide distributions within the monazite concentrate.

Figure 4: TREO basket composition (Source: see Appendix 5; price data as per average REO prices during LOM)

NEAR-ZERO INCREMENTAL OPERATING COSTS

Kasiya's REC will be recovered from mineral streams the DFS flowsheet already produces. Upstream steps including free-dig mining, ore transport to the plant, the scrubber, wet concentration and electrostatic separation have been defined and costed in the DFS and accordingly attributed to rutile and graphite production. The only incremental processing is the added monazite concentrate circuit which involves spiral gravity separation and flotation treating the non-conductor stream from the Mineral Separation Plant, together with product packaging and storage. There is no additional mining, no early-stage processing and no change to the DFS mining method, wet concentration or graphite recovery strategy.

Incremental operating costs are accordingly low. The incremental site operating cost is ~US$0.90/kg REC at the mine gate. Including transport and port charges, the cost is ~US$1.39/kg REC free-on-board FOB Dar es Salaam. Including ocean freight and insurance, the total cost is ~US$3.68/kg REC delivered CIF Houston, Texas. The Study deliberately costs Kasiya's REC delivered into the United States. On that basis, the operating margin is approximately 90%. Due to its by-product nature, this cost structure holds across rare earth price cycles. Including all other regulatory fees, Sovereign can land REC in America for US$3.85/kg in any market.

Figure 5: Kasiya's circuit to produce REC is added onto the DFS operation

PROFITABLE EVEN AT U.S. GOVERNMENT FLOOR PRICES

Since Sovereign's REC may be a by-product of the titanium and graphite operation defined in the DFS, Kasiya REC supply does not depend on rare earth prices to remain in production. This is reflected in the Study's two main price cases: the Base Case (using Argus Media price forecasts) delivers the potential for an incremental pre-tax NPV₈ of US$722 million at a 151% IRR, and the Western Supply Case (reflecting higher demand for ex-China supply of monazite concentrate) US$883 million at 172%.

Figure 6: Kasiya pre-tax NPV and IRR in various market and geopolitical scenarios

A third case tests the bottom of the market as the U.S. Government itself has defined it. Since July 2025, guaranteed minimum prices have become a standard feature of U.S. rare earth supply arrangements:

• a US$110/kg floor for NdPr established with MP Materials and repeated with Lynas; and
• the first Western floor prices for heavy rare earths, at US$575/kg for Dy and US$2,050/kg for Tb

The U.S. Floor Price Case applies these floors, with 2025 average prices for yttrium, samarium and gadolinium which do not currently have floor prices but are deemed critical. Even in this scenario, the circuit generates US$183 million of incremental pre-tax NPV₈ and a 43% IRR. These floor prices arise under U.S. Government arrangements with third parties and are applied as a downside assumption only; Kasiya is not party to any such arrangement.

Figure 7: Effect of China's heavy rare earths export controls on Tb and Dy prices in US$/kg (vs. floor price and long-term forecast prices per Study)

A SOLUTION TO CHINA EXPORT CONTROLS AND SUPPLY DOMINANCE

In April 2025, China introduced export controls over seven medium and heavy rare earths - including dysprosium, terbium, yttrium, samarium and gadolinium, each present in Kasiya's concentrate - citing their dual-use applications. Dysprosium, terbium and yttrium together comprise approximately 15% of Sovereign's TREO basket. In October 2025, China added five more REEs - holmium, erbium, thulium, europium and ytterbium.

Figure 8: Geopolitical importance of Kasiya's REC vs. current non-Chinese REE producers

(Source: Appendix 5)

China accounts for most global mined production of these elements and effectively all commercial separation capacity. The United States is 100% net import-reliant on yttrium, with nearly all supply derived from concentrates processed in China (Source: USGS Mineral Commodity Summaries, 2026). The U.S. defence industry is estimated to hold a two-to-three-year inventory of samarium with potential replenishment of munitions expended in the 2025 Iran conflict adding to demand.

FEEDSTOCK FOR RARE EARTH REFINING CAPACITY FUNDED BY THE U.S. AND ALLIES

The Western rare earth supply chain comprises four stages: primary mining and concentrate production; separation into individual rare earth oxides; conversion to metals and alloys; and magnet manufacturing. Western governments and industry are investing across all four. Availability of heavy rare earth feedstock is a key choke point determining whether the downstream links can operate at capacity.

Western primary production is dominated by light rare earth deposits, with only trace quantities of dysprosium, terbium and yttrium. New separation capacity is being financed, commissioned and constructed specifically to process heavy rare earths:

• the United States (e.g. MP Materials Corp. with funding from the Department of War (DOW));
• Australia (e.g. Iluka Resources Limited with funding from Export Finance Australia (EFA); Tronox Holdings plc with funding from Export-Import Bank of the United States (EXIM) and EFA); and
• Malaysia (Lynas Rare Earths Limited with funding from the DOW)

This new capacity requires a qualified, scalable supply of heavy rare earth feed. Downstream, metal and alloy producers and magnet manufacturers depend on separated dysprosium and terbium for the high-temperature NdFeB magnets used in defence, aerospace, electric vehicle and robotics applications.

Kasiya's REC product contains approximately 1,485tpa of TREO over an initial 23-year mine life and carries five of the seven rare earths placed under Chinese export controls in April 2025.

The REC is uncommitted, at a time when Western governments have established price floors, government-backed offtake vehicles and bilateral supply frameworks specifically to secure heavy rare earth feed from allied jurisdictions. Kasiya is positioned to supply that feed into Western separation capacity for the duration of its mine life.

Figure 9: Kasiya's role in the Western rare earth supply chain

CRITICAL ELEMENTS FOR CRITICAL PLATFORMS

Production of heavy rare earths outside China remains negligible yet various critical platforms of the West depend on at least one rare earth Kasiya's REC will contain.

This dependence occurs throughout defence, energy, technology and medical systems. NdFeB permanent magnets - built on NdPr, with dysprosium and terbium added wherever they must perform at high temperature - sit in the actuators, motors and generators of fighter aircraft, guided missiles, submarines, drones, electric vehicles and humanoid robots. Yttrium hardens jet engines and semiconductor plasma-etch chambers and enables military radar and solid-state lasers; samarium-cobalt magnets guide missiles; gadolinium serves naval propulsion, turbine coatings and the world's MRI machines. Publicly reported estimates indicate approximately 410kg of rare earth materials in each F-35-class airframe, approximately 4.2 tonnes in each Virginia-class submarine, and 1-4kg in each electric vehicle or humanoid robot.

Every platform shown in Figure 10 depends on at least one rare earth contained within Kasiya's REC product.

Figure 10: Importance of Heavy Rare Earths to the West

NEXT STEPS

• Variability testwork: confirm monazite recoveries and concentrate quality across the DFS mine schedule
• Product qualification: REC samples from Sovereign's Lilongwe facilities to prospective customers for evaluation
• Marketing: offtake discussions with Western processors and government procurement programmes
• Pre-Feasibility Study: rare earths integrated into Kasiya's definitive development case; completion targeted 2027
• Resource development: Assessment of whether there can be Inferred conversion and extension of the monazite MRE beyond the 25-year DFS pits

The April 2026 DFS established Kasiya as a large-scale, long-life natural rutile and natural flake graphite development in central Malawi. The DFS is based on a 25-year mine life, phased throughput of 12 Mtpa in Phase 1 expanding to 24 Mtpa in Phase 2, and 535.8 Mt of Ore Reserve scheduled for processing. At steady state, the DFS forecasts approximately 222 ktpa of natural rutile and 275 ktpa of natural flake graphite, with a pre-tax NPV8 of US$2.204 billion and a pre-tax IRR of 23.4%.

The upstream DFS battery limit is the non-conductor stream from the Mineral Separation Plant (MSP). The incremental scoping study scope principally covers downstream separation of this stream into an REC, product packaging, storage, and logistics. It does not assess a separate mining operation and does not modify the April 2026 Kasiya Ore Reserve, mine schedule, Phase 1 and Phase 2 throughput strategy or primary rutile and graphite production plan.

Basis of the production target

The production profile for the monazite product is derived from the April 2026 DFS production schedule and the characteristics of the MSP stream, together with Scoping Study-level assumptions for recovery, yield, product grade, ramp-up and availability.

• The production target is based on the non-conductor streams generated in the DFS production schedule, the characterisation of those streams with respect to monazite and recovery assumptions are based on metallurgical testwork and benchmarking.
• The Monazite Mineral Resource Estimate, covering the April 2026 DFS life of mine pit shells of which 74% is in the Indicated Category, supports the Monazite bearing non-conductor feed into the REC recovery circuit.
• The study assumes the Kasiya DFS project is funded, constructed and operated substantially as scheduled, such that this scoping study is conducted on an incremental basis to produce the REC product.
• No separate or additional Ore Reserve is being declared for purposes of this scoping study.

TABLE 3: STUDY CONTRIBUTORS

KASIYA DFS CONTEXT

April 2026 DFS Reference

Kasiya is located approximately 30 km north-west of Lilongwe, Malawi. The April 2026 DFS defines a phased, owner-operated, shallow free-dig development producing premium natural rutile and natural flake graphite for export via the Nacala Corridor. Tailings are progressively returned to mined-out pits, avoiding a conventional surface tailings storage facility.

This Scoping Study is incremental to the April 2026 Kasiya DFS and does not contemplate a separate mine development. It retains the existing DFS mining and primary-processing basis and assesses the additional downstream facilities and activities required to recover and produce a REC from the non-conductor mineral streams generated within the Kasiya MSP.

The April 2026 DFS basis retained includes:

• the existing Ore Reserve and mine schedule
• the dry-mining method and pit sequence
• Phase 1 and Phase 2 plant throughput
• the primary wet-concentration circuit and rutile-graphite production strategy
• the use of existing site infrastructure and utilities as designed.

The incremental scope assessed in the Scoping Study includes:

• additional treatment of the non-conductor stream to recover and produce a REC
• additional mineral separation and product finishing facilities
• incremental infrastructure tie-ins and utility requirements
• product storage, packaging and containerisation
• product handling, logistics and marketing assumptions
• incremental CAPEX, OPEX, permitting and funding requirements.

All capital costs, operating costs, revenues and financial outcomes reported for the monazite product are assessed on an incremental basis relative to the April 2026 Kasiya DFS base case.

TABLE 4: APRIL 2026 DFS REFERENCE PARAMETERS

GEOLOGY AND RESOURCES

The Kasiya deposit is located on the Lilongwe Plain in central Malawi and is hosted by deeply weathered paragneiss and orthogneiss of the Mozambique Belt Basement Complex. The mineralised system forms broad, relatively flat-lying and laterally continuous bodies within a well-developed weathering profile. Rutile is preferentially enriched in the near-surface horizons, while graphite grades generally increase below approximately six metres. Monazite containing elevated concentrations of the heavy rare earth elements dysprosium, terbium and yttrium has also been recovered from the rutile processing circuit, indicating the potential for an additional rare REC product.

The mineralised zones extend for more than 10 kilometres along strike and are approximately one to four kilometres wide. Drilling has been completed on regional 800-metre centres, with progressive infill to 400 metres and 200 metres in key areas. Geological interpretation and continuity have been supported by twin drilling, close-spaced geostatistical drilling, trial-mining channel samples and open-pit sampling. The existing rutile and graphite Mineral Resource Estimate is classified as Measured, Indicated and Inferred in accordance with the JORC Code (2012).

Monazite

Monazite mineralisation occurs within the same deeply weathered, near-surface residual heavy-mineral system that hosts the Kasiya rutile and graphite Mineral Resource. The minerals form part of the naturally occurring heavy-mineral assemblage derived from the weathering of the underlying paragneiss and are distributed throughout the extensive, sub-horizontal mineralised profile. Monazite is an important source of light rare earth elements, including neodymium and praseodymium, while monazite-associated rare earth minerals are enriched in heavy rare earth elements.

Monazite Mineral Resource Estimate

Sovereign has advanced a comprehensive analytical programme to support an additional Mineral Resource estimate for monazite within the April 2026 DFS mine plan. A total of 3,250 magnetic heavy-mineral composites, representing 1,012 boreholes within the DFS pit areas, were submitted for XRF analysis. Selected composites were also submitted for full-suite rare earth element analysis by ICP to assess changes in rare earth composition across the deposit and through the weathering profile.

The analytical and statistical work indicates that monazite occurs with consistent rare earth assemblage, although relative proportions vary with depth and weathering. For resource estimation purposes, lanthanum oxide has been used as the principal analytical proxy and cerium oxide as a supporting validation measure.

The by-product MRE for monazite in-situ mineral percentage constrained to the DFS open pits for the Kasiya rutile and graphite project is 524.4 Mt at 0.0132% Monazite. The MRE is classified as 74% Indicated and 26% Inferred.

TABLE 5: KASIYA 2026 MONAZITE MINERAL RESOURCE ESTIMATE

Notes:

• The mineral resource is constrained to the Kasiya rutile and graphite DFS open pits to satisfy RPEEE
• No cut-off is applied and the total volume within the DFS open pits will be processed with monazite concentrate being produced as a by-product of the Kasiya rutile and graphite production
• The grade % is the in-situ grade of monazite mineral calculated from the recovered magnetic concentrate fraction of the HMS within the sand fraction (>45um <600um)

MINING

Kasiya is planned as a large-scale, conventional dry-mining operation exploiting soft, free-dig material to a maximum depth of approximately 20 metres. Pilot mining demonstrated that the weathered ore can be excavated without drilling, blasting, crushing, grinding or milling. The selected mining method uses draglines equipped with approximately 12.7 cubic metre buckets to load 100-tonne rigid dump trucks, which transport run-of-mine material to the processing plants or designated stockpiles.

Mining will generally be undertaken in two benches: a five-metre upper cut and a lower cut of up to 15 metres. This configuration enables early access to higher-grade near-surface rutile, assists with management of the plant-feed graphite grade and keeps the dragline above the groundwater table, which typically occurs at depths of seven to nine metres. Mining is scheduled to commence before processing to establish ore stockpiles and create void capacity for subsequent in-pit tailings deposition.

Development is based on two 12 Mtpa processing modules. The South operation commences first, followed by the North operation from Year 5, increasing total nameplate capacity to 24 Mtpa. Plant locations were selected near the respective centres of the resource to reduce haulage distances and facilitate progressive backfilling. Recovery of monazite as a REC would use material already mined and delivered for rutile and graphite production.

The Scoping Study is based on the same pit shells and mine plan as the DFS excluding any years in which Inferred Resources account for >50% of the mining inventory i.e. years 24 and 25. It incorporates DFS-level assumptions for the dry-mining method, mining costs, recoveries, exclusion zones and operating constraints. These assumptions have been validated by trial mining and reviewed by the relevant Competent Person. The rare earth recovery circuit does not require additional mining, changes to the Ore Reserve, changes to the mining sequence or additional run-of-mine material.

Figure 12: Monazite plant feed over LOM

Sovereign confirms that the potential incremental recovery of monazite from mineral streams the subject of the Scoping Study is financially viable when excluding Inferred Mineral Resources in the production schedule. Also, and as identified above in Figure 12, the Inferred Mineral Resources do not feature as a significant proportion early in the Scoping Study mine plan.

PROCESSING AND METALLURGY

Metallurgy testwork and design basis

The Scoping Study is supported by metallurgical testwork undertaken on Kasiya material, supplemented by relevant benchmark data for the Scoping Study level design. The test work procedure is summarised below.

Rare Earth Concentrate

• Kasiya bulk metallurgical samples previously processed at Allied Mineral Laboratories were utilised.
• Non-conductor concentrates generated from these samples were retrieved from storage for processing.
• QEMSCAN, an automated microscope-based analysis, was used to identify the minerals present and measure the monazite content of each non-conductor concentrate.
• Electron Probe Microanalysis (EPMA) was used to determine the chemical composition of individual monazite particles.
• Scientific Services in Cape Town analysed the concentrates using X-ray Fluorescence (XRF) to measure the major elements present and Inductively Coupled Plasma (ICP) analysis to measure specific elements, including those present in very small quantities.
• Sovereign's laboratory in Malawi used size, gravity, electrostatic and magnetic separation techniques to produce a REC with elevated levels of heavy rare earth oxides in the TREO content.

The design of the gravity circuit and gravity circuit recovery assumptions were informed by the metallurgical testwork programme. Monazite flotation benchmark data was used to inform the flotation circuit design.

Further variability, flowsheet optimisation and locked-cycle testwork will be required in the next study phase to confirm design criteria and product performance.

Please refer to announcements dated 21 January 2026 and 27 May 2026 for further details.

Process design

The Scoping Study process concept to recover the additional minerals is simply an add-on to the Kasiya MSP as defined in the DFS. There is no change to the DFS mining method, wet concentration and graphite recovery strategy.

The non-conductor stream from the MSP passes through the added monazite circuit. This circuit consists of feed preparation and a range of unit operations (including spiral gravity separation and flotation) to produce a monazite concentrate.

Monazite concentrate then reports to the product drumming facilities. The design philosophy uses conventional mineral sands and monazite separation technology, and maximises practical reuse of existing utilities and infrastructure.

Figure 13: REC production and shipping

Process metrics

Recovery and product assumptions are supported by the metallurgical programme described above.

Monazite feed grade - The non-conductor grades of monazite and % mass reporting to the non-conductor stream were used to back calculate the resource feed grade. The study applies 80% recovery to monazite through the spirals and flotation circuit adopted for the Base Case.

TABLE 6: PROCESS METRICS

Production Profile

Recovery and production of the REC follows the phased ramp-up of the existing Kasiya project. Phase 1 is based on 12 Mtpa of DFS plant throughput and Phase 2 reflects the expansion to 24 Mtpa. The incremental product production profile is derived from the non-conductor streams generated under the DFS production schedule, supported by characterisation of its monazite grade and Scoping Study-level metallurgical recovery and processing assumptions.

TABLE 7: PRODUCTION PROFILE

PRODUCT SPECIFICATIONS

REC product

This scoping study builds on the existing DFS adding a REC. The Kasiya Rare REC is dominantly made up of the mineral monazite and also contains minor xenotime (typical of commercial monazite concentrate mine gate specification) plus numerous gangue minerals. As monazite concentrate is a common product term, this concentrate is referred to as both monazite and REC in the text of this document.

Neodymium oxide (Nd₂O₃), representing 16.6% by weight, and praseodymium oxide (Pr₆O₁₁), at 4.2%, are especially significant. Both are essential feedstocks for high-strength neodymium-iron-boron permanent magnets used in electric-vehicle motors, wind-turbine generators, industrial robotics, consumer electronics and defence systems. Their importance to electrification and renewable-energy technologies makes them major drivers of the product's commercial value.

Dysprosium oxide (Dy₂O₃), present at 2.1%, and terbium oxide (Tb₄O₇), at 0.4%, represents highly strategic components. These elements improve the heat resistance and magnetic stability of permanent magnets, enabling reliable operation in demanding applications such as electric drivetrains, aerospace equipment and advanced military systems. Terbium is also used in phosphors, specialised electronics and magneto strictive materials.

TABLE 8: KASIYA REC TREO DISTRIBUTION

The concentrate contains several other useful rare earth oxides. Yttrium oxide (Y₂O₃), at 13.0%, is widely used in advanced ceramics, thermal-barrier coatings, lasers, phosphors and high-performance alloys. Gadolinium oxide (Gd₂O₃) has commercial applications in medical-imaging contrast agents, neutron-absorbing materials, nuclear technology and specialised magnetic products. Samarium oxide (Sm₂O₃), present at 2.9%, is used in samarium-cobalt permanent magnets valued for their high-temperature performance, corrosion resistance and reliability in aerospace, defence and industrial systems.

Together, these oxides give the product relevance across clean energy, transport, electronics, healthcare, nuclear technology and national defence. The combination of magnet-related rare earths-particularly neodymium, praseodymium, dysprosium and terbium - strengthens its strategic importance as a potential feedstock for downstream separation, refining and advanced-material manufacturing.

TRANSPORT AND LOGISTICS

Route, packaging and cost assumptions

Dar es Salaam has been selected as the base-case export route for the Kasiya REC product, with onward shipping to the United States. The current transport concept involves road haulage of sealed drums in 20-foot containers, with one container per truck, using managed convoys supported by emergency-response capabilities and security escorts as required. The plan incorporates journey management, real-time tracking and route-risk assessments. Matters requiring further confirmation include the final product classification, Class 7 packaging and package payload, licensing, customs, border and port requirements in Malawi and Tanzania, shipping availability, container cycles, export and import documentation and potential demurrage. The final production profile, shipment frequency, route-specific logistics costs and port handling and shipping arrangements will be confirmed during the next study phase.

TABLE 9: TRANSPORT AND LOGISTICS PARAMETERS

RARE EARTH MARKET

TABLE 10: MARKET OUTLOOK FOR KEY ELEMENTS WITHIN KASIYA'S REC

Figure 14: Rare Earth Demand by Element (2026-2040)

(Source: Argus)

Global rare earth demand is forecast to increase from 222,500 tonnes of rare earth oxide (REO) in 2026 to approximately 345,000 tonnes by 2040, representing a compound annual growth rate (CAGR) of 3.2%. This growth is expected to be led by permanent magnets, where demand is forecast to increase by 115% over the same period, at a CAGR of 5.6%.

Permanent magnets are expected to be the principal driver of rare earth demand, substantially outpacing the forecast annual growth of 1.4% across all other applications combined. Key sources of demand include:

• Electric vehicles: Permanent-magnet motors remain widely used in mass-market EVs due to their efficiency, power density and contribution to vehicle range. Although some manufacturers are assessing magnet-light or magnet-free alternatives, performance, weight and cost trade-offs may limit widespread substitution. Continued growth in global EV adoption is therefore expected to support magnet demand through the 2030s.
• Offshore wind: The expansion of offshore wind generation is expected to support demand for large permanent magnets, particularly in direct-drive turbines. Offshore installations are forecast to grow faster than onshore capacity, supported by higher capacity factors and fewer land constraints.
• Advanced robotics: The emergence of humanoid and mobile-service robots could become an additional source of demand from the mid-2030s. These systems typically require numerous compact, precise and energy-efficient actuators - characteristics well suited to permanent-magnet technologies.

Demand from established applications, including glass polishing, ceramics and metallurgical additives, is expected to grow broadly in line with global industrial activity. Cerium-based polishing products and additives should continue to benefit from demand in construction, consumer electronics and precision optics, although improvements in material efficiency may moderate growth.

Pricing

The market assessment uses independent Argus information on rare earth supply, demand and pricing. The financial model applies an oxide basket price derived from the forecast composition of the Kasiya concentrate, Argus price assumptions for the individual rare earth oxides and the adopted TREO content and payability assumptions. The assessment recognises the concentrate's exposure to NdPr and to the supply-constrained heavy and specialised rare earths dysprosium, terbium, yttrium, samarium and gadolinium.

The Study assumes that a saleable heavy REC can be produced and sold into established rare earth processing markets. Product specification, payability, impurity limits, customer qualification, radionuclide acceptance and final commercial terms remain subject to confirmatory testwork, customer engagement and marketing studies in the next phase.

The financial model uses a REO basket price derived from the forecast oxide composition of the REC product and individual rare earth oxide price assumptions.

Payable basket price = REO weights x REO prices x TREO % x Payability %

The REC basket price for Kasiya is calculated using REO prices provided by Argus , applying the contained TREO %, and Argus' recommended payability factor.

COSTS ESTIMATIONS

Capital Costs (Capex)

The incremental pre-production capital expenditure (CAPEX) estimate was developed by DRA.

The estimate has been estimated to Scoping Study accuracy of approximately ±30% and includes the monazite separation circuit, product storage and packaging facilities, as well as incremental utilities, associated buildings, Class 7 handling provisions, EPCM/owners' costs and contingency. The total incremental pre-production CAPEX is US$57 million, comprising:

TABLE 11: PROJECT CAPEX OVER LOM (US$)

The capital cost estimate base date is Q3 2026. Sustaining Capex over LoM is US$20m.

Operating Costs (Opex)

Sovereign's forecasts for the operating cost has been estimated at a Scoping Study accuracy of ±30% for the production of an REC at Kasiya over the life-of-mine.

TABLE 12: OPERATING COSTS (US$/KG)

The REC operating cost estimate has been prepared on an incremental basis relative to the April 2026 Kasiya DFS. Accordingly, it includes only the additional costs associated with recovering, handling and transporting the REC product; the existing mining and upstream rutile-graphite processing cost base remains within the DFS. Total incremental operating costs (FOB Dar es Salaam) are estimated at approximately US$74.9 million over the mine life, comprising US$47.1 million of processing costs, US$1.7 million of additional site administration costs and US$26.1 million of off-site costs.

Processing represents approximately 63% of total incremental FOB operating costs. The main components are stores and consumables, including flotation reagents, plant maintenance, product drums and dryer fuel; power consumption for the two REC plants; and additional labour. The operating structure remains relatively lean, with an estimated 19 additional employees during Phase 1, increasing to 38 employees following the Phase 2 expansion.

Off-site costs account for approximately 35% of total FOB operating costs and principally comprise Class 7-compliant road transport to Dar es Salaam and associated port charges. These allowances reflect the specialist packaging, handling, documentation and transport requirements applicable to a concentrate containing naturally occurring thorium and uranium. The estimate has been developed at Scoping Study level and will be refined through further metallurgical testwork, logistics studies, product classification and supplier quotations during the next phase of work

FINANCIAL AND ECONOMIC ANALYSIS

Modelling Assumptions

The economic evaluation is undertaken on an incremental basis relative to the April 2026 Kasiya DFS base case. Only capital, operating costs and revenues attributable to the additional monazite opportunity are included, except where an allocation of existing infrastructure cost is explicitly stated. The model is presented at a discount rate of 8% and a study period aligned with the 25 year DFS mine life. However, years 24 and 25 are excluded from any economic evaluation since plant feed in these years includes more than 50% inferred resources. Accordingly, the life of mine for this Study is 23-years.

Key Economic Outcomes

The base case demonstrates a potentially material incremental value contribution to Kasiya under the adopted technical, market and cost assumptions. Key outcomes are summarised below.

TABLE 13: KEY ECONOMIC OUTCOMES

Sensitivity Analysis

Sensitivity analysis assesses the impact of changes in REE prices, incremental CAPEX, site OPEX and selling/logistics costs on pre-tax NPV. The results indicate that project value is most sensitive to the REE basket price and discount rate applied, with the base case remaining positive under the selected downside scenario, including the Floor Price Scenario as detailed in this announcement.

TABLE 14: NPV (PRE-TAX) SENSITIVITY TO KEY INPUTS

ENVIRONMENTAL AND SOCIAL

The Environmental and Social Impact Assessment and Environmental, Social and Governance workstreams are supported by multi-season baseline studies, specialist environmental and social studies, trial mining, soil-rehabilitation programmes, agricultural livelihood-restoration programmes and development of an Environmental and Social Management System aligned with the IFC Performance Standards. These workstreams have received substantial review and input from Rio Tinto social and environmental subject-matter experts through strategic workshops, site visits, stakeholder engagement, review of specialist studies and management plans, regular technical sessions and in-country support.

The Study provides for secure, controlled-access storage and handling areas, compliant sealed packaging, appropriate monitoring, trained personnel, real-time shipment tracking, security measures, and comprehensive transport, emergency-response and incident-management procedures.

LEGAL AND TENURE

The Company owns 100% of the relevant Kasiya exploration and retention licences through its wholly owned Malawi-registered subsidiaries. The tenements are in good standing and the established pathway for conversion to mining tenure and project permitting applies to the integrated Kasiya development. No known legal, tenure or third-party constraint has been identified that would materially prevent development of the Project. Applicable government, vendor and community-development royalties and obligations are included in the economic evaluation where relevant.

The rare earth scope introduces additional requirements relating to Class 7 product storage, packaging and transport, cross-border movement, customs, port handling and export documentation. These requirements are incorporated at Scoping Study level and will require final legal, regulatory and specialist confirmation before operations commence.

Following the completion of this DFS in April 2026, Sovereign applied for a Mining Licence (ML) to secure mineral deposits for mining and includes the licence areas covered by the DFS and this Scoping Study. The ML application specifically includes rare earth oxides, which will provide comprehensive coverage for the Project's mineral suite once the ML is awarded. At this point of Kasiya's development, the Company notes no known issues or impediments obtaining a ML under normal course of business.

Under the 2023 Mines Act, the Government of Malawi has a right to equity ownership for large-scale mining licences (>5Mt mined per annum or >US$250m Capex) with the right a negotiation matter, likely as part of any future MDA. The Mkango and Lotus MDAs (refer details below in the modifying factors) included a 10% non-diluting equity interest to the Malawi Government.

RISKS

Key project risks identified and mitigation measures implemented were as follows.

In addition, strategic and external risks, including macroeconomic factors, commodity price volatility, and supply chain constraints, were assessed and incorporated into the financial evaluation and ongoing project planning.

Risk management will continue throughout the project lifecycle, with the risk register maintained as a live document and subject to regular review and update as the project progresses.

SUMMARY OF MATERIAL ASSUMPTIONS

The material assumptions underpinning the production target and forecast financial information are summarised below. The production profile for REC is supported by the April 2026 DFS production schedule (excluding years with >50% of plant feed in Inferred category of Mineral Resources) together with metallurgical testwork recovery, product yield and MSP by-product stream information.

TABLE 16: MATERIAL ASSUMPTIONS

DISCLOSURES & DISCLAIMERS

Competent Person Statements

Monazite Scoping Study

The information in this announcement that relates to the Mineral Resource Estimate (Monazite) is based on, and fairly represents, information compiled by Mr Jeremy Witley, a Competent Person (and qualified person) who is a member of the South African Council for Natural Scientific Professions (SACNASP Pr. Sci. Nat.), a Recognised Professional Organisation (RPO) included in a list promulgated by ASX from time to time. Mr Witley is a principal of MSA Group, an independent consulting company. Mr Witley has sufficient experience relevant to the style of mineralisation and type of deposit under consideration, and to the activity he is undertaking, to qualify as a Competent Person as defined in the 2012 Edition of the 'Australasian Code for Reporting of Exploration Results, Mineral Resources and Ore Reserves' (JORC Code) and a Qualified Person under the AIM Rules for Companies. Mr Witley consents to the inclusion of the matters based on his information in the form and context in which they appear.

The information in this announcement that relates to Exploration Results (Monazite) is and fairly represents, information compiled by Mr Malcolm Titley, a Competent Person (and qualified person) who is a member of The Australasian Institute of Mining and Metallurgy and member of Australian Institute of Geoscientists. Mr Titley is the Technical Services Manager of Sovereign and is a holder of shares and unlisted performance rights in Sovereign. Mr Titley has sufficient experience that is relevant to the style of mineralisation and type of deposit under consideration and to the activity being undertaken, to qualify as a Competent Person as defined in the JORC Code' and a Qualified Person under the AIM Rules for Companies. Mr Titley consents to the inclusion in the announcement of the matters based on his information in the form and context in which it appears.

The information in this announcement that relates to Production Targets is based on, and fairly represents, information provided by Mr Frikkie Fourie, a Competent Person (and qualified person), who is an Associate Member of The South African Institute of Mining and Metallurgy and a Registered Professional Engineer with the Engineering Council of South Africa, a RPO, included in a list promulgated by ASX from time to time. Mr Fourie is employed by Moletech Consulting Pty Ltd, an independent consulting company. Mr Fourie has sufficient experience, which is relevant to the style of mineralisation and type of deposit under consideration, and to the activities undertaken, to qualify as a Competent Person as defined in the JORC Code and a Qualified Person under the AIM Rules for Companies. Mr Fourie consents to the inclusion in the Announcement of the matters based on his information in the form and context in which it appears.

The information in this announcement that relates to metallurgical testwork and performance, monazite processing facilities and operating cost estimate is based on, and fairly represents, information provided by Mr Johann Ludwig Niebuhr, a Competent Person (and qualified person), who is a Registered Professional Engineer with the Engineering Council of South Africa, a RPO, included in a list promulgated by ASX from time to time. Mr Niebuhr is employed by DRA Projects (Pty) Ltd, an independent consulting company. Mr Niebuhr has sufficient experience, which is relevant to the style of mineralisation and type of deposit under consideration, and to the activities undertaken, to qualify as a Competent Person as defined in the JORC Code and a Qualified Person under the AIM Rules for Companies. Mr Niebuhr consents to the inclusion in the Announcement of the matters based on his information in the form and context in which it appears.

The information in this announcement that relates to the Scoping Study (including infrastructure and capital costs) is based on, and fairly represents, information provided by Mr Robin Mark Welsh, a Competent Person (and qualified person), who is a Registered Professional Engineer with the Engineering Council of South Africa, a RPO, included in a list promulgated by ASX from time to time. Mr Welsh is employed by DRA Projects (Pty) Ltd, an independent consulting company. Mr Welsh has sufficient experience, which is relevant to the style of mineralisation and type of deposit under consideration, and to the activities undertaken, to qualify as a Competent Person as defined in the JORC Code and a Qualified Person under the AIM Rules for Companies. Mr Welsh consents to the inclusion in the Announcement of the matters based on his information in the form and context in which it appears.

The information in this announcement that relates to the economic analysis is based on, and fairly represents, information provided by Ruan Albert Daffue, a Competent Person (and qualified person), who is a member of the South African Institute of Mining and Metallurgy, a RPO, included in a list promulgated by ASX from time to time. Mr Daffue is employed by Practara (Pty) Ltd, an independent consulting company. Mr Daffue has sufficient experience, which is relevant to the style of mineralisation and type of deposit under consideration, and to the activities undertaken, to qualify as a Competent Person as defined in the JORC Code and a Qualified Person under the AIM Rules for Companies. Mr Daffue consents to the inclusion in the Announcement of the matters based on his information in the form and context in which it appears.

Definitive Feasibility Study

The information in this announcement that relates to the DFS (including Mine Engineering, Mine Scheduling, Processing, Infrastructure, Capital and Operating Costs, Production Target and Ore Reserves and other financial forecasts) is extracted from an announcement dated 16 April 2026, which is available to view at www.sovereignmetals.com.au.

Sovereign confirms that a) it is not aware of any new information or data that materially affects the information included in the original announcement; b) all material assumptions included in the original announcement continue to apply and have not materially changed; and c) the form and context in which the relevant Competent Persons' findings are presented in this announcement have not been materially changed from the original announcement.

The information in this announcement that relates to the exploration results (metallurgy) is extracted from announcements dated 28 September 2023, 8 May 2024, 15 May 2024, 4 September 2024, 21 January 2026 and 27 May 2026, which is available to view at www.sovereignmetals.com.au. Sovereign confirms that a) it is not aware of any new information or data that materially affects the information included in the original announcement; b) all material assumptions included in the original announcement continue to apply and have not materially changed; and c) the form and context in which the relevant Competent Persons' findings are presented in this announcement have not been materially changed from the original announcement.

TABLE 17: KASIYA ORE RESERVE - RUTILE (RUT95) + GRAPHITE (TGC)

The Total Ore Reserve is all rutile and graphite mineralisation within an optimised open pit shell using a Rut95 concentrate revenue price of net US$1,286.81/t and a Graphite product price of net US$1,099.51/t; Mine Opex US$1.35/t; Process Opex US$5.44/t; Rutile recovery of 97.6%; Average Graphite recovery of 70.4%.

Modifying Factors

The Modifying Factors as defined under the JORC Code, including mining, processing, metallurgical, infrastructure, economic, marketing, legal, environmental, social and governmental factors, have been assessed as part of the Study and are reflected in the Study Announcement.

A summary assessment of each relevant Modifying Factor is provided below.

Mining - unchanged to the DFS

Refer to section entitled 'Mining' in the Announcement.

Approximately 99% of the scheduled throughput (production target) of this Scoping Study for the first twelve and half years of estimated production is in the Indicated Mineral Resource category with only 1% in the Inferred Mineral Resource category. For the life of mine of this Scoping Study, approximately 72% of life of mine production target is in the Indicated Mineral Resource category and 28% is in the Inferred Mineral Resource category.

The Company has concluded that it has reasonable grounds for disclosing a production target which includes an amount of Inferred material on the basis that the Inferred Resources included in the early mine plan is modest and over the life of the mine the amount of Inferred Resources is not the determining factor in project viability.

However, there is a low level of geological confidence associated with Inferred mineral resources and there is no certainty that further exploration work on the Kasiya deposit will result in the determination of additional Indicated Mineral Resources or that the production target (or financial forecasts) itself will be realised.

In the unlikely event that the remaining Inferred Mineral Resources are not able to be upgraded, the Project's viability is not affected. This is supported by a stand-alone discounted cash flow analysis prepared that assumes only Indicated Resources are included the mine plan in order to demonstrate the economic viability of the Project.

Metallurgy and Processing - incremental update to the DFS

Refer to section entitled 'Processing and Metallurgy' in the Announcement

The upstream processing basis is unchanged from the DFS. Extensive variability and pilot-scale testwork was completed across the rutile and graphite streams to support DFS-level flowsheets and mass balances, confirming the ability to produce premium-specification rutile and graphite products using conventional processing routes.

The Scoping Study assesses an additional circuit downstream of the MSP to treat the existing non-conductor stream and recover a REC containing monazite. The selected flowsheet comprises feed preparation, spiral gravity separation, flotation and final gravity cleaning, followed by product dewatering, handling and drumming. The design uses conventional mineral-sands separation technology and is supported by Kasiya-specific gravity testwork, mineralogical and chemical characterisation, flotation benchmarking and comparable operating and development-project flowsheets. The Study applies 80% recovery to both monazite and assumes production of a concentrate containing 90% combined monazite and approximately 56%-56.6% TREO. Further representative variability, flowsheet optimisation and locked-cycle testwork, together with final product and radionuclide characterisation, will be completed in the next study phase.

Infrastructure - incremental update to the DFS

Refer to sections entitled 'Transport and Logistics' and 'Cost Estimations' in the Announcement

DFS-level engineering has been completed for power, water, roads, buildings, laboratory capability, load-out facilities, transport and logistics infrastructure, including access to hydropower, established rail corridors and port infrastructure. The infrastructure design is aligned with the mine plan, plant locations and staged development from 12Mtpa to 24Mtpa.

The incremental rare earth scope includes the additional recovery and product-finishing plant, utility tie-ins, product storage and drumming facilities, Class 7-compliant handling and storage provisions, and associated buildings and services. Existing DFS infrastructure and operating systems are reused where practicable. The base-case export concept is road transport of sealed drums in 20-foot containers from Kasiya to the Port of Dar es Salaam, with managed convoy movements, journey management, real-time tracking, emergency-response capability and security escort as required. Final route, port acceptance, package design, licensing, shipping availability and costs remain subject to specialist confirmation.

Marketing - additional marketing for REC

Refer to section entitled 'Rare Earth Market' in the Announcement

The market assessment uses independent Argus information on rare earth supply, demand and pricing. The financial model applies an oxide basket price derived from the forecast composition of the Kasiya concentrate, Argus price assumptions for the individual rare earth oxides and the adopted TREO content and payability assumptions. The assessment recognises the concentrate's exposure to NdPr and to the supply-constrained heavy and specialised rare earths dysprosium, terbium, yttrium, samarium and gadolinium.

The Study assumes that a saleable REC can be produced and sold into established rare earth processing markets. Product specification, payability, impurity limits, customer qualification, radionuclide acceptance and final commercial terms remain subject to confirmatory testwork, customer engagement and marketing studies in the next phase.

Environmental, Social, Legal and Governmental - unchanged to the DFS

Refer to section entitled 'Environmental and Social' in the Announcement

The Environmental and Social Impact Assessment and Environmental, Social and Governance workstreams are supported by multi-season baseline studies, specialist environmental and social studies, trial mining, soil-rehabilitation programmes, agricultural livelihood-restoration programmes and development of an Environmental and Social Management System aligned with the IFC Performance Standards. These workstreams have received substantial review and input from Rio Tinto social and environmental subject-matter experts through strategic workshops, site visits, stakeholder engagement, review of specialist studies and management plans, regular technical sessions and in-country support.

The tailings-management strategy uses integrated in-pit backfilling aligned with the life-of-mine mining schedule and avoids a conventional external tailings storage facility. Tailings are deposited as flocculated co-disposal using engineered slurry handling, pumping, pipeline, drainage, decant and stormwater-control systems. The mining sequence maintains void space for backfilling and incorporates a conservative 12-month consolidation period before traffic accesses backfilled areas. This approach reduces the Project footprint, lowers resettlement requirements and supports progressive rehabilitation and water management.

The REC is expected to contain naturally occurring thorium and uranium and is assumed, for the Scoping Study, to require Class 7 controls. The Study includes allowance for controlled-access storage and handling areas, sealed and compliant packaging, monitoring and dosimetry as required, trained personnel, journey and emergency-response planning, real-time tracking, security controls and incident-management procedures. Final radiological risk assessment, activity-concentration testing, package classification and design, management plans and stakeholder engagement will be completed as the Project advances.

Legal and Tenure

The Company owns 100% of the relevant Kasiya exploration and retention licences through its wholly owned Malawi-registered subsidiaries. The tenements are in good standing and the established pathway for conversion to mining tenure and project permitting applies to the integrated Kasiya development. No known legal, tenure or third-party constraint has been identified that would materially prevent development of the Project. Applicable government, vendor and community-development royalties and obligations are included in the economic evaluation where relevant.

The rare earth scope introduces additional requirements relating to Class 7 product storage, packaging and transport, cross-border movement, customs, port handling and export documentation. These requirements are incorporated at Scoping Study level and will require final legal, regulatory and specialist confirmation before operations commence.

Following the completion of this DFS in April 2026, Sovereign applied for a Mining Licence (ML) to secure mineral deposits for mining and includes the licence areas covered by the DFS and this Scoping Study. The ML application specifically include rare earth oxides, which will provide comprehensive coverage for the Project's mineral suite once the ML is awarded. At this point of Kasiya's development, the Company notes no known issues or impediments obtaining a ML under normal course of business.

Under the 2023 Mines Act, the Government of Malawi has a right to equity ownership for large-scale mining licences (>5Mt mined per annum or >US$250m Capex) with the right a negotiation matter, likely as part of any future MDA. The Mkango and Lotus MDAs (refer details below) included a 10% non-diluting equity interest to the Malawi Government.

Economic

Refer to sections entitled 'Cost Estimations' and 'Financial and Economic Analysis' in the Announcement

Practara Metals and Mining Advisory developed the incremental rare earth financial model. The evaluation retains the DFS mine schedule and primary rutile and graphite operating basis and includes only the additional capital costs, operating costs, product-handling and logistics costs, revenues, royalties and other cash flows attributable to the rare earth opportunity, except where use of shared infrastructure is expressly allocated. The model is prepared in real US dollars, uses the Study's 23-year production period and applies the disclosed discount rate, recovery, ramp-up, availability, product, pricing and payability assumptions.

Capital inputs comprise the additional separation and finishing circuit, utility tie-ins, product storage and packaging, Class 7 provisions, associated buildings, engineering and owners' costs and contingency. Operating inputs include labour, power, reagents and consumables, maintenance, product handling, packaging and off-site logistics. The model supports the reported incremental economic outcomes and has been tested for changes in rare earth prices, capital costs, site operating costs and selling and logistics costs. As a Scoping Study, the estimates and outcomes remain subject to the accuracy range, qualifications and risks disclosed in this announcement and to refinement through further technical and commercial work.

A Government royalty of 5% (applied to revenue) and a vendor profit share of 2% (applied to gross profit) have been included in all project economics. A 0.45% royalty (applied to revenue) has been applied for the community development fund.

The DCF analysis demonstrated compelling economics of the prospective by-product, with an NPV (ungeared, pre-tax, at an 8% discount rate) of US$722 million, and an (ungeared, pre-tax) IRR of 151%.

Sensitivity analysis was performed on all key assumptions used. The robust project economics insulate the Kasiya Project from variations in market pricing, capital expense, or operating expenses. Specifically, the Floor Price Case as defined in this announcement demonstrates the REC by-product's economic viability in a commodity price downside scenario. The monazite Project breakeven REC basket price is ~US$4,140/t which would require a ~90% decline in forecast REE prices used in the Study. Based on the Shanghai Metals Market, the current spot price for basic ~55% monazite concentrate is ~US$6,400/t in China.

Further, a sensitivity analysis with the entire REC basket price 25% lower than the Scoping Study selected prices, the Project still displays a positive NPV (ungeared, pre-tax, 8% discount rate) of US$521 million and pre-tax IRR of 122%.

The Project's REC by-product payback period is 1.5 years from the start of production. The payback period is based on unlevered, pre-tax free cashflow.

Sovereign estimates the incremental capital cost to construct the monazite circuit for first production to be US$29m (which includes a contingency of US$3.8 million).

Key parameters are disclosed in the body of the announcement, and include:

• Life of Mine: 23 years
• Discount rate: 8%
• Royalty rate: 5% royalty (Government), 2% of gross profit (Original Project Vendor) and 0.45% Community Development Fund.
• Pricing: Rare earth oxide price forecasts as per Argus with Base Case at 50% payability for a 56.6% TREO grade concentrate

There is uncertainty in some respects of the tax law applicable to mining companies in Malawi. Specifically with regard to the calculation of and the application of the RRT. The 2023 Mines Act, further provides for the holder of large-scale mining licenses to enter into a fiscal stability agreement, known as a Mine Development Agreement (MDA) with the Government of Malawi. These MDAs provide the opportunity to agree and clarify the application of taxation. At the date of the DFS announcement, Sovereign has not applied for a large-scale mining license or entered into an MDA.

In 2024, mining companies Lotus Resources Limited (Lotus) and Mkango Resources Limited (Mkango) each entered into separate MDAs with the Government of Malawi. These MDAs have deviated from the enacted mining taxation laws and, specifically, the levying of the RRT, with no RRT payable as part of the MDAs. The Government has since proposed an alternative supernormal profits tax (SPT) to replace the current RRT. Profits of up to MWK 10 billion (~US$6 million at the time of writing) are taxed at the standard income tax rate of 30%. Any profits which exceed MWK 10 billion are taxed at a higher rate of 40%. Mkango and Lotus are exempt of SPT under their existing MDAs. As such, and until such time Sovereign has sight of what actual fiscal terms would apply to the Kasiya Project in terms of its own fiscal stability agreement with the Government of Malawi, results for the DFS have been reported on a pre-tax basis only.

Notwithstanding this uncertainty, scenario analysis has been undertaken to indicate a range of potential post-tax outcomes. This analysis considered the impact of key fiscal variables, including RRT, SPT, and capital allowances, on project returns, with the resulting post-tax NPV8% estimated to range between US$458 million and US$535 million.

The financial model has been built and prepared by Practara using inputs from the various expert consultants and has been reviewed by SP Angel Corporate Finance LLP (SPA), the Company's Nominated Advisor and Corporate Broker as defined by the AIM Rules for Companies set out by the London Stock Exchange, to validate the functionality and accuracy of the model.

The Company considers that the required funding for incremental capex costs to first production could be raised through an equity placement within its available 15% placement capacity under ASX Listing Rule 7.1. This assessment is supported by Sovereign's demonstrated access to equity capital markets. Sovereign also has a strong track record of successfully raising equity funds for Kasiya plus it has an experienced and high-quality Board and management team comprising highly respected resource executives with extensive technical, financial, commercial and capital markets experience. The directors have previously raised more than A$2.5 billion from capital markets for a number of exploration and development companies.

Further, the Company has also had preliminary financing discussions with a number of other institutional investors, development finance institutions, U.S. and "Western-aligned" government agencies, end-user customers, and other strategic investors, regarding potential equity and/or debt funding at the Company and/or Project level.

The Company considers that, given the nature of the Project, funding is likely to be sourced from specialist investors. Potential funding sources include, but are not limited to, traditional equity and debt, royalty financing, and off-take agreements at either the corporate and/or Project level.

Since commencing exploration at Kasiya in November 2019, the Company has completed extensive drilling, sampling, metallurgical test work, and geological modelling, and has defined a Measured and Indicated Mineral Resource Estimate that has been converted to Ore Reserves as part of this DFS. The Company is also in a unique position, having collected real-world data through the Pilot Mining Program, which has significantly validated and de-risked the Project. Over this period, and with these key milestones being achieved, the Company's market capitalisation has increased from approximately A$18m to over A$330m.

The Company has a simple and clean corporate and capital structure, is debt-free, and is in a strong financial position, with approximately A$25.1 million in cash on hand as at 30 June 2026 (unaudited). This financial position means the Company is well funded to continue key further project workstreams, including the permitting and financing activities required to advance the Project to the next phase of study's, engineering and final investment decision.

The Company's shares are listed on the ASX and AIM which are premier markets for growth companies and provide increased access to capital from institutional and retail investors in Australia and the UK. The Company's shares are also quoted on the OTCQX and Frankfurt Stock Exchange.

The Company has also entered into multiple agreements that may provide alternative sources of financing, as set out below:

· A Collaboration Agreement with the International Finance Corporation (IFC), a member of the World Bank Group, which provides a clear pathway to international project financing for Kasiya, with a particular focus on debt funding. Through the Collaboration Agreement, IFC holds rights to participate as lender, mandated co-lead arranger and/or investor in the project financing structure, supporting the development of a robust and bankable funding package. As a leading global development finance institution with extensive experience in mining projects in emerging markets, IFC's involvement is expected to enhance lender confidence, support access to competitive long-term debt, and underpin the overall financing strategy for the Project; and

• A non-binding MOU with a PE fund to fund the development and construction of a 132kV transmission line connecting Kasiya to the Malawi national grid. Refer to discussion above in Infrastructure Modifying Factor.

As discussed above, and taking into account the following additional factors: 1) Recently completed funding arrangements for similar or larger scale development projects; 2) The range of potential funding options available; 3) The favourable key metrics generated by the Kasiya Project; 4) Investor interest to date; 5) The Company owns 100% of Kasiya which is highly attractive to potential financiers; and 6) The previously completed DFS demonstrated that the Project is commercially viable and provides justification to progress to the final investment decision and project finance stages, the Company has a high level of confidence that the Project will be able to secure funding in due course.

APPENDIX 1- SUMMARY OF RESOURCE ESTIMATE REPORTING CRITERIA

As per ASX Listing Rule 5.8 and the JORC reporting guidelines, a summary of the material information used to estimate the MRE is detailed below.

Geology

Regional Geology

The greater part of Malawi is underlain by crystalline Precambrian to lower Palaeozoic rocks referred to as the Malawi Basement Complex. In some parts, these rocks have been overlain unconformably by sedimentary and volcanic rocks ranging in age from Permo-Triassic to Quaternary. The Basement complex has undergone a prolonged structural and metamorphic history dominated by uplift and faulting, resulting in the formation of the Malawi Rift Valley.

Kasiya is located on the Lilongwe Plain, which is underlain by the Basement Complex paragneisses and orthogneisses, which are part of the Mozambique Belt. The bulk of the gneisses are semi-pelitic, but there are bands of psammitic and calcareous rocks that have been metamorphosed under high pressure and temperature conditions to granulite facies.

Interspersed within the paragneiss units are lesser orthogneisses, often cropping out as conspicuous tors, as well as amphibolites, pegmatites and minor mafic to ultramafic intrusions. Foliation and banding in the gneisses have a broad north-south strike over the general area. Thick residual soils and pedolith with some alluvium overlie the gneisses and include sandy, lateritic and dambo types.

Project Geology

Sovereign's tenure covers 644 km2 over an area to the north, west and south of Malawi's capital city, covering the Lilongwe Plain. The topography is generally flat to gently undulating, and the underlying geology is dominated by paragneiss with pelitic, psammitic and calcareous units.

A particular paragneiss unit is rich in rutile and graphite and is the primary source of both minerals in the area. The same geological unit contains monazite with associated rare earth elements which will be mined as a by-product of the rutile and graphite operation. This area was deeply weathered during the Tertiary, and rutile concentrated in the upper part of the weathering profile, forming residual placers, such as the Kasiya deposit. Once this material is incised and eroded, it is transported and deposited into wide, regional braided river systems, forming alluvial heavy mineral placers such as the Bua Channel.

Kasiya Deposit Geology

The high-grade rutile deposit at Kasiya is best described as a residual placer, or otherwise known as an eluvial heavy mineral deposit. It is formed by weathering of the primary host rock and concentration in place of heavy minerals, as opposed to the high-energy transport and concentration of heavy minerals in a traditional placer.

The presence of abundant kyanite and graphite in the host material suggests a meta-sedimentary protolith. The protolith likely started with a 0.5-1.5Ga basin that also experienced a consistent influx of titanium minerals.

These sedimentary rocks were subject to granulite facies metamorphism under reduced conditions in the Pan-African Orogeny. The metamorphic facies, reduced environment, relatively high titanium content and low iron content resulted in rutile being the most stable titanium mineral under these conditions. Slow exhumation and cooling then resulted in re-crystallisation as paragneisses containing coarse rutile, graphite and monazite.

The final and most important stage of rutile enrichment came as tropical weathering during the Tertiary depleted the top ~8m of physically and chemically mobile minerals. This caused significant volume loss and concurrent concentration of heavy resistate minerals, including rutile, monazite, ilmenite and kyanite.

Rutile and monazite mineralisation therefore lies in laterally extensive, near-surface, flat "blanket" style bodies in areas where the weathering profile is preserved.

Monazite mineralisation shows no clear relationship with either rutile or graphite mineralisation. Indications at this stage are that grades are a function of both weathering and geological variations in a vertical and lateral sense respectively. The results from previous work shows an enrichment in Y2O3 in samples from the upper weathering units compared to that of samples representing the lower weathering units. This is interpreted as a change in the relative abundance of monazite and the primary yttrium-bearing mineral, with the latter enriched in the upper weathering units due to its greater resistance to weathering compared with monazite.

Monazite mineralisation follows distinct lateral orientations indicating that it is determined to some degree by varying geological units within the Kasiya deposit.

Drilling Techniques

Spiral hand-auger (HA) drilling, Push-tube and/or diamond core (PTDD), and Air core (AC) drilling methods have been used extensively at the Kasiya deposit by Sovereign to define mineralisation and to obtain quantitative rutile, monazite and graphite assay information.

HA drilling was executed by Sovereign field teams using a manually operated enclosed-flight Spiral Auger (SP / SOS) system produced by Dormer Engineering in Queensland, Australia. The HA bits are 62mm and 75mm in diameter with 1m long steel rods. Each 1m of drill advance is withdrawn and the contents of the auger flight removed into bags and set aside. An additional 1m steel rod is attached and the open hole is re-entered to drill the next metre. This is repeated until the drill hole is terminated often due to the water table being reached or due to bit refusal. The auger bits and flights are cleaned between each metre of sampling to avoid contamination.

PTDD drilling is undertaken using a drop hammer Dando Terrier MK1 and a drop hammer DL650 by Geo-consult and Thompsons Drilling. The drilling generated 1m runs of 88mm PQ core in the first 2m and then transition to 61mm core for the remainder of the hole. Core drilling is oriented vertically by spirit level.

AC drilling was completed by Thompson Drilling utilising a Smith Capital 10R3H compact track-mounted drill. The drilling is vertical and generates 1m samples with care taken in the top metres to ensure good recoveries of the high-grade surface material. The AC sample is collected by the on-board cyclone into heavy-duty RC sample bags. Drilling continues until bit refusal onto basement ~20-30m. Sample bags are immediately transported back to Sovereign's field laydown yard where they are processed. AC drilling is on a nominal 200m by 200m pattern.

The PTDD and AC twin and density sample holes are selectively placed throughout the deposit to ensure a broad geographical and lithological coverage for the analysis.

MSA has reviewed Standard Operating Procedures (SOPs) for HA, SA, PTDD and AC drilling and found them to be fit for purpose and support the resource classifications as applied to the MRE.

Sampling Techniques

HA samples are obtained at 1m intervals generating on average approximately 2.5kg of drill sample. HA samples are manually removed from the auger bit and sample recovery is visually assessed in the field. As samples become wet at the water table and recovery per metre declines, the drill hole is terminated. Each 1m sample is sun dried, logged and weighed. HA samples are composited based on regolith weathering boundaries defined by geology logging. Each 1m of sample is dried, lightly pressed to remove soft aggregates and riffle-split to generate a total sample weight of 3kg for analysis, generally at 2 - 5m intervals. This primary sample is then riffle split again to provide a 1.5kg sample each for rutile + monazite and graphite analyses.

SA samples are bulk spiral auger samples primarily designed to collect a large sample for metallurgical and pilot plant testwork. Bit sizes range from 300 mm to 700mm diameter. The samples are collected on 1m intervals, laid out on a large tarpaulin to be sun dried before using a cone and quarter method (for the 700mm diameter) to produce a roughly 100kg sample which is then riffle split to produce a 3kg sample, with the file split providing 1.5 kg each for rutile and graphite analysis. Samples are analysed in 1m intervals.

PTDD samples are predominantly from HQ sized core (61mm diameter). Half core 1m samples are sun dried, logged and weighed. Samples are then lightly pressed and composited over 2m intervals. An equal mass is taken from each contributing metre to generate a 1.5kg composite sample. Individual recoveries of core samples are recorded on a quantitative basis. Core recovery is very good overall at >95%.

AC samples are collected in 1m increments. AC samples are dried, riffle split, lightly pressed and composited. Samples are collected and homogenised prior to splitting to ensure sample representivity. ~1.5kg composite samples were defined by the regolith boundaries in earlier drilling. More recent AC drilling utilised regular 2m downhole composites. An equivalent mass is taken from each primary sample to make up the composite.

During 2024 twin drilling campaigns, samples were processed at 1m intervals to get a better understanding of drilling and deposit variability.

The sampling and compositing methods described are considered appropriate and reliable based on accepted industry practice. MSA completed an on-site audit of sampling and sample processing and deemed the processes fit for purpose.

Sample Analysis Methodology

Monazite grade was determined by XRF analysis of the magnetic heavy mineral (HM) fractions that were produced during the processing of drill samples for rutile assessment. The rutile sample analysis methodology has been reported on in the past but is again included for completeness.

All samples arrive at Sovereign's Malawi laboratory where they are sorted and checked in. Graphite samples are identified and prepared for export, while the equivalent rutile and monazite samples begin the sample workflow to generate the rutile non-magnetic concentrate (NMag) for export for TiO2 and multi-element XRF analysis and the monazite magnetic concentrate (Mag) for monazite and REE analysis. Prior to June 2024 XRF analysis was completed at ALS Perth, Western Australia, currently Scientific Services South Africa (SS) laboratory in Cape Town, South Africa is being used. Umpire checks have shown good correlation between the two external laboratories. Audit of Sovereign's laboratory premises, staff, sample analysis and QA procedures was completed by MSA during two site visits in 2024 and 2025.

SVM Malawi Laboratory Rutile Workflow

• Samples are dried in a commercial oven for 1 hour at 105℃ and a dry raw samples mass is recorded.
• Samples are soaked in 1% Tetrasodium pyrophosphate (TSPP) solution overnight and then lightly agitated prior to wet screening.
• Wet screening occurs at 5mm, 600µm and 45µm to remove oversize and slimes (-45µm) material. Each +45µm retained fraction is dried, logged and weighed.
• The resulting SAND fraction +45µm -600mm is oven dried for 1 hour at 105℃ after which its dry weight is recorded.
• The SAND fraction is then passed over a Gemeni wet shaking table at a constant feed rate to generate a heavy mineral concentrate (HMC).
• Heavy Liquid Separation (HLS) at Diamantina Laboratories in Perth was initially trialled as a preferred separation method but was quickly superseded (supported by QA analysis) by wet-table separation on account of substantial near-density gangue material reporting to the HM sink for the HLS technique. The HLS analyses represent 6% of the MRE assay dataset.
• The wet-tabled HMC is then subject to magnetic separation @ 16,800G (2.9Amps), producing the Mag and NMag concentrate fractions. The separation is performed using a Mineral Technologies Reading Pilot IRM (Induced Roll Magnetic) purchased by Sovereign and located at the Company's laboratory in Malawi. Pre-2022, this step was completed by Allied Mineral Laboratories Perth (AML) in Perth, Western Australia.

The Malawi onsite laboratory sample preparation methods are considered quantitative to the point where the Mag (containing monazite) and NMag concentrate (containing the rutile) are produced.

Recovered monazite is determined by using a multiplication factor of 8.3 to convert La2O3 to a combined monazite product and is defined and reported here as: monazite recovered in the SAND +45 to -600um range to the Mag concentrate fraction as a % of the total primary, dry, raw sample mass. i.e recoverable monazite mineral assemblage. Monazite are treated as a single REE-bearing mineral assemblage.

QAQC

Accuracy monitoring is achieved through submission of certified reference materials (CRM's). Sovereign uses internal and externally sourced wet screening reference material inserted into samples batches at a rate of 1 in 20.

For the mag XRF analysis SS use internal CRMs and duplicates on XRF analyses.

Analysis of sample duplicates is undertaken by standard statistical methodologies (Scatter, Pair Difference and QQ Plots) to test for bias and to ensure that sample splitting is representative. Standards determine assay accuracy performance, monitored on control charts, where failure (beyond 2SD from the accepted mean value of the standard) initiates investigation and may trigger re-processing of the affected batch.

Examination of the QA/QC sample data indicates satisfactory performance of field sampling protocols and assay laboratories providing acceptable levels of precision and accuracy.

Magnetic Fraction Selection and Compositing Methodology

Magnetic HM fractions for XRF analysis were selected from boreholes within and on the margins around the pits scheduled in the Kasiya DFS production plan. The magnetic fractions represent intervals based on the weathering units and boundaries interpreted for the Kasiya rutile and graphite mineral resource.

The magnetic fractions from the selected boreholes were composited further within a weathering unit for two reasons:

• The preparation of a fused bead for XRF analysis requires a minimum of 4g of material and where the weight of the magnetic fraction was inadequate, material from adjacent intervals within the same weathering unit were combined to produce the required weight.
• As a time and cost saving measure adjacent magnetic fractions were composited within weathering boundaries to minimise the number of samples sent for XRF analysis. Mining at Kasiya will be done as separate cuts on upper and lower weathering units which aligns with compositing the magnetic fraction to represent weathering units for resource modelling purposes.

Magnetic fractions were composited proportionally so that the contribution of each component is represented in the combined sample by weighted average of bulk.

Monazite Content Determination Methodology

The content of the monazite mineral within the magnetic fraction is determined by multiplying the La2O3 content by a factor of 8.3. The multiplication factor was derived from the XRF and ICP results from several borehole and metallurgical monazite rich samples across the Kasiya MRE summarised below and in Table 18.

• 0m to 6m and 6m to 20m composited magnetic fractions from two AC boreholes from the Sparrow and Kingfisher pits
• Nine electrostatic non-conductor concentrates produced from bulk samples taken from 0m to 6m and 6m to 20m intervals from pit and AC material in Kingfisher, Mousebird and Sparrow.
• 2 electrostatic non-conductor products from two ROM bulk samples from Babbler.
• Non-conductor products from the nine DFS bulk ROM samples from the southern and northern pits.

TABLE 18: MONAZITE CHEMICAL COMPOSITION FROM XRF AND ICP RESULTS

Monazite chemical formula for the above samples were estimated from the ICP and XRF results by adding the individual REO, P2O5, U3O8 and ThO2. The content of each REO within the monazite mineral assembly was calculated. La2O3 showed the least variation accounting for 12% of the monazite mineral with a STD of 0.66 compared to that of CeO2 accounting for 23% with a STD Dev of 1.87. A multiplication factor of 8.3 was subsequently used to estimate monazite from La2O3 content of magnetic fraction XRF analysis. Using a 4.3 multiplication factor to calculate monazite for CeO2 produced the same average result, but with increased variability in the near surface material.

Monazite in-situ grade was then calculated for the composite by using the magnetic fraction as percentage of bulk material and the monazite content within the magnetic fraction.

Estimation Methodology

Datamine Studio RM with supporting statistical software was used for the data analysis, variography, geological interpretation and mineral resource estimation.

A 3D block model honouring the geology boundaries which included weathering horizons; barren mafic intrusives; surface clay horizons and presence of amphibolite was created. The model was also coded with the tenement EL codes, DFS 25 year production open pit boundaries, rock in-situ dry bulk density and moisture content.

Eight grade domains (2 in each weathering zone) were created, 4 mineralised and 4 low grade for monazite. The domains are derived from the combination of weathering type inside or outside the mineralisation DTM's. Samples were composited to 1 sample per drillhole per domain.

The composite populations generally approximated log normal distributions with some -ve and/or +ve skewness relating to the imposed mineralisation boundaries.

Ordinary Kriging (OK) was considered the best grade estimator for monazite due to the near log normal grade populations and suitable variograms. Variography analysis was used to determine domain nugget effect and OK search and neighbourhood parameters.

Each grade domain was treated as a 2D seam and estimated using OK with dynamic anisotropy which followed the same broad geology protolith continuity trends as used for the rutile and graphite MRE. No declustering or removal of twin data was required, as OK is an efficient declustering algorithm, and the post OK checks demonstrated very few negative weights in the mineralised zones. Any areas not estimated due to inadequate sample coverage were set to null and excluded from the MRE.

The parent cell size used is roughly equivalent to the average drill hole spacing within the Indicated Resource (200m*200m). XY sub-celling to 50m*50m is adequate resolution for horizontal boundaries. Seam modelling ensured the mineralisation, weathering and topography layers were vertically accurate (within the 50m horizontal resolution). Grade was estimated using the parent cell panel size.

Grade estimation was constrained by hard boundaries (domains) that result from the geological interpretation and mineralisation interpretation.

Top Capping was applied to the composites considered to be outliers to reduce local high grade bias. Generally <1% of samples had a grade cap applied.

Validation of the grade estimate was completed both visually and statistically. Visual validation by loading the model and drill hole files and annotating, colouring and using filtering to check for the appropriateness of the estimate. Distributions of section line averages (swath plots) for drill holes and models were prepared for each zone and orientation for comparison purposes.

The resource model has appropriately utilised the informing drill hole data and is considered suitable to support the resource classifications applied to the estimate.

In-situ dry bulk density was calculated from 400 core samples taken from geographically and lithologically representative sites across the deposit. Dry bulk density is calculated from PT drill core using a cylinder volume wet and dry method performed by Sovereign in Malawi. Shelby tube core samples collected from the 2024 PTDD drill program were analysed by CIVILAB in South Africa.

Bulk density data was coded by weathering horizon. Population distributions were then reviewed and obvious outlies removed. Either the mean or median were used as the average for each weathering and/or rock type domain.

The average in-situ dry bulk density of the total MRE is 1.60 t/m3. This is derived from using an average density of 1.39 t/m3 for the SOIL; 1.58 t/m3 for the FERP, 1.66 t/m3 for the MOTT; 1.68 t/m3 for the PSAP; and 1.77 t/m3 for SAPL. The bulk density is unchanged from the MRE used in the DFS.

Mining and Metallurgy Factors

Dry-mining has been determined as the optimal method of mining for the Kasiya Rutile deposit. The materials competence is loose, soft, fine and friable with no cemented sand or dense clay layers, allowing for a free dig mining method. It is considered that the strip ratio would be zero or near zero. Dilution is not a consideration for monazite, as all the material within the designed DFS 25 year open pits will be bulk processed through the process plant.

Metallurgical test work and benchmarking provided monazite recovery data.

Nine bulk metallurgical testwork samples processed at Allied Mineral Laboratories (AML) though previous scoping, prefeasibility and definitive feasibility studies. These samples provided a starting point for metallurgical testwork for monazite.

Each of the samples processed at AML generated a non-conductor tailing sample from the electrostatic circuit within the MSP. This stream contains most of the monazite. The nine samples were subject to further analysis to determine the mineralogy of the non-conductor stream, and chemical analysis of the non-conductor stream. Chemical analysis of the monazite minerals within the non-conductor stream was also performed.

Mineralogy was determined by QEMscan, monazite chemistry by Electon Probe Microanalysis (EPMA), non-conductor elemental assay by ICP and XRF. Analysis was undertaken at the University of Cape Town (UCT) for the Qemscan and EPMA and Scientific Services (SS) for the XRF and ICP.

As at August 2026 metallurgical test work has been limited to gravity and magnetic separation at Sovereign Services Limited Lilongwe laboratory (SSL). This was performed on a non-conductor sample produced from HMC processed at Light Deep Earths Johannesburg. This testwork provided the monazite recovery from the non-conductor stream to the monazite concentrate. Flotation recovery was benchmarked from flotation testwork performed on monazite samples from mineral sands deposits. Reference data was provided by Mineral Technologies.

A recovery of 80% is applied to monazite processed through the monazite gravity and flotation circuit which accepts feed from the MSP non-conductor stream.

Classification

The Kasiya MRE has been classified as Indicated or Inferred. No material has been classified as Measured due to the use of La2O3 as a proxy to estimate monazite concentration which is still undergoing additional fine tuning, with additional ICP and EPMA analysis.

JORC classification considered geological understanding; mineralisation continuity; drilling and sampling quality and spacing; OK estimation efficiency (KE) and confidence (SoR); with consideration of the proposed mining method and scale.

The dominant control on monazite grade distribution within the mineralised zone is intensity of weathering, with slightly lower average grades with higher variance in the upper more intensely weathered soil and FERP, with increasing grade continuity through MOTT into PSAP and SAPL. Low grade zones are likely related to changes in the protolith lithology.

Monazite mineralisation has been well defined by drilling with appropriate sample analysis to determine in-situ monazite grade. The mineralisation zone is generally broad and continuous with mineralisation grade affected either by changes in the protolith or displaced by mafic intrusives. Recent drainage has also impacted mineralisation continuity. Minor near surface clay lenses and metamorphic 'pegmatitic' zones also displace mineralisation. These very minor internal 'waste' zones are visually identifiable during mining (as seen during the 2024 trial mining exercise) and can be selectively either mined or bypassed.

Regional exploration was completed on a nominal 800m square grid, with infill to 400m followed by either 200m square or 200m offset grid. Twin holes have demonstrated the robustness of the geology interpretation and mineralisation continuity.

KE generally exceeds 0.4 up to 0.8 in areas with detailed infill drilling. SoR generally exceeds 0.5 with most areas exceeding 0.7 where drilling is around 400m x 200m or closer.

On the basis of the high confidence geology interpretation; mineralisation scale and continuity including taking into account the bulk mining method; and very tight grade distributions within the estimation domains, the Competent Person is comfortable classifying all of the monazite mineralisation which lies above the base of drilling as either Indicated or Inferred.

Indicated was defined using a nominal KE >=0.3 to 0.5 and a SOR >=0.5, which generally matches areas with a nominal drill spacing of 200 to 400m. A boundary was used to define both the Indicated and Inferred Mineral Resource.

The monazite MRE is reported with the DFS 25 year plan Kasiya rutile and graphite open pit shells to reflect the JORC Code requirement for Reasonable Prospects for Eventual Economic Extraction (RPEEE). The open pit shells are the results of a full DFS assessment and financial modelling based on the DFS mine production plan.

The MRE is presented in Table 5.

Cuf-off grades

No cut-off grade is applied for estimation of the monazite mineral resource as all material within the DFS mine plan will be processed through the Kasiya processing plant, regardless of monazite grade.

APPENDIX 2 - JORC CODE, 2012 EDITION - TABLE 1

Section 1 - Sampling Techniques and Data

Two similar designs of HA drilling equipment are employed. HA drilling with 75mm diameter enclosed spiral bits (SOS) with 1m long steel rods and with 62mm diameter open spiral bits (SP) with 1m long steel rods. The SP bit accounts for less than 10% of the HA drilling, as the enclosed spiral proved to be the more effective tool. Drilling is oriented vertically by eye. Each 1m of drill sample is collected into separate sample bags and set aside. The auger bits and flights are cleaned between each metre of sampling to avoid contamination. Core-drilling is undertaken using a drop hammer, Dando Terrier MK1. The drilling generated 1m runs of 88mm PQ core in the first 2m and then transitioned to 61mm core for the remainder of the hole. Core drilling is oriented vertically by spirit level. AC drilling was completed by Thompson Drilling utilising a Smith Capital 10R3H compact track-mounted drill. Each 1m sample bag is immediately transported back to Sovereign's secure field laydown yard for processing.

Dry +45µm -600mm (sand fraction) in oven for 1 hour at 105 ℃ Pass +45µm -600mm (sand fraction) across wet table to generate a HMC. Dry HMC in oven for 30 minutes at 105 ℃ Magnetic separation of the HMC by Carpco magnet @ 16,800G (2.9Amps) into a magnetic (Mag) and non-magnetic (NMag) fraction Send NMag to external laboratory for TiO2% (and other elements) XRF analysis Send the Mag to external laboratory for La2O3, CeO2, Y2O3 and Nd2O3% (and other elements) XRF analysis Various workflows were used to produce HMC, Magnetic separation and external laboratory TiO2% plus other XRF analysis prior to June 2023. All magnetic fractions from samples generated in Lilongwe and sent to Perth based laboratories or generated at Perth based laboratories prior to 2023 were shipped to Scientific Services in Cape town for XRF analysis. Work flow codes and number of samples impacted are presented below: WORKFLOW Num Sample Metres DIA-AML-IT 190 635.0 DIA-AML-ALS 877 2,860.2 LLW-AML-IT 408 1,465.5 LLW-AML-ALS 3,321 8,745.8 LLW-LLW-ALS 5,272 9,279.3 LLW-LLW-SS 7,768 12,959.2 Total 17,836 35,944.9 DIA-AML-IT and DIA-AML-ALS The Sand fractions are sent to Diamantina Laboratories, Perth. Split ~150g of sand fraction for HLS using Tetrabromoethane (TBE, SG 2.96g/cc) as the liquid heavy media to generate HMC. (Heavy liquid separation (HLS) of the HM is no longer required and a HM result is not reported in the updated MRE. The HMC prepared via wet-table, gravity separation at the Lilongwe Laboratory provides an ideal sample for subsequent magnetic separation and XRF.) Bag the HMC fraction and send to AML Perth for quantitative separation. The resulting NM fractions are sent to either ALS Metallurgy Perth or Intertek Perth for quantitative XRF analysis. LLW-AML-IT and LLW-AML-ALS Bag HMC fraction and send to Perth, Australia for quantitative separation at AML The resulting NM fractions are sent to either ALS Metallurgy Perth or Intertek Perth for quantitative XRF analysis. LLW-LLW-ALS The NM fractions are sent to ALS Metallurgy Perth for quantitative XRF analysis. Samples receive XRF_MS and are analysed for: TiO2, Al2O3, CaO, Cr2O3, Fe2O3, K2O, MgO, MnO, SiO2, V2O5, ZrO2, HfO2. LLW-LLW-SS The NM fractions are sent to Scientific Services South Africa for quantitative XRF analysis. Samples are analysed for: TiO2, Nd2O3, CeO2, La2O3, BaO, HfO2, Nb2O5, ZrO2, Y2O3, Fe2O3, MnO, Cr2O3, V2O5, CaO, K2O, P2O5, SiO2, Al2O3, MgO, NaO2 The Mag fractions are sent to Scientific Services South Africa for quantitative XRF analysis. Samples are analysed for: TiO2, Nd2O3, CeO2, La2O3, BaO, HfO2, Nb2O5, ZrO2, Y2O3, Fe2O3, MnO, Cr2O3, V2O5, CaO, K2O, P2O5, SiO2, Al2O3, MgO, NaO2 The number of Monazite samples used in the MRE by hole type are: HTYPE_4 Num Sample Metres AC 1,087 7,209.33 HA 645 2,425.30 PTDD 474 1,958.98 SA 203 931.80 Total 2,389 12,525.41

REE CRMs for XRF on magnetic HM fractions were inserted by Scientific Services. Examination of the QA/QC sample data indicates satisfactory performance of field sampling protocols and assay laboratories providing acceptable levels of precision and accuracy. Acceptable levels of accuracy and precision are displayed in statistical analyses to support the resource classifications as applied to the estimate.

Further laboratory and metallurgical analysis of the Mag fraction to improve the definition of the Rare Earth Elements (REE) associated with the presence of Monazite in the Mag fraction. Further analysis of other potential HM associated with the byproducts of rutile production, including the Non-Mag fraction.

APPENDIX 3 - MONAZITE PLAN VIEW AND ASSAY RESULTS

Figure 15: Plan view showing the Kasiya mine plan pits with the drillholes sampled for monazite

APPENDIX 4 - SOURCE DATA

U.S. rare earth production, consumption and import reliance

U.S. Geological Survey, Mineral Commodity Summaries 2026 (February 2026)

• U.S. rare earth production of 8,900t REO-equivalent in 2025 (4,300t in 2024);
• net import reliance approximately 67%;
• yttrium among 13 minerals at 100% net import reliance, with China the dominant supplier;
• no U.S. production of heavy rare earths (including Dy, Tb) at commercial scale;
• U.S. mineral-reliant industries US$4.09 trillion of output, more than one-eighth of the U.S. economy
• Yttrium consumption basis (~550tpa) per USGS yttrium statistics, 2021-2025 average.
• Chinese export controls

Ministry of Commerce of the People's Republic of China: export controls on seven medium and heavy rare earths (samarium, gadolinium, terbium, dysprosium, lutetium, scandium, yttrium), April 2025; controls extended to holmium, erbium, thulium, europium and ytterbium, October 2025.

• U.S. Government floor prices and supply arrangements
• MP Materials / U.S. Department of Defense partnership announcement (10 July 2025): ten-year US$110/kg NdPr floor price. (LINK)
• Lynas Rare Earths ASX announcement (16 March 2026): U.S. Department of War supply arrangement applying the US$110/kg NdPr floor.(LINK)
• U.S. Department of War / Serra Verde offtake structure (20 April 2026): US$1.55 billion government-backed vehicle; floor prices of US$575/kg dysprosium oxide and US$2,050/kg terbium oxide, escalating 2% per annum under a 15-year take-or-pay arrangement. (LINK)
• Samarium inventories and defence content

National Defense Magazine, "Stockpiles of Critical Mineral Needed for High-End Magnets Begin to Dwindle" (1 July 2026)

Platform rare earth content

U.S. Congressional Research Service / U.S. Department of Defense figures as publicly reported: approximately 410kg of rare earth materials per F-35 airframe; approximately 4.2t per Virginia-class submarine.

Adamas Intelligence: Per-unit content of electric vehicles and humanoid robots (~1-4kg)

End-use equivalents - bases of Sovereign's derivations

~7 million humanoid robots: contained NdPr + DyTb over the initial 23-year mine life at per-robot magnet content per item 5 above

≈18% of U.S. Dy/Tb separation nameplate: 36tpa contained DyTb against the 200tpa nameplate of the first U.S. heavy rare earth separation circuit (MP Materials, Q3 2025 results)

≈35% of U.S. yttrium consumption: 193tpa Y₂O₃ against average annual U.S. consumption per item 1.

≈70% of the world's MRI gadolinium: ~43tpa contained Gd against >50t of gadolinium administered annually in MRI worldwide (ACR/NKF consensus statement, Radiology, 2021: ~50 million doses per year; Davies et al., BioMetals, 2022, citing Caravan et al.)

APPENDIX 5 - PEER TREO DISTRIBUTION DATA