Magnes Rare Earths’ First RC Results Return 29 m at 0.58% TREO to 60 m Depth; Auger Hole Returns 15 m From Surface at 1.15% TREO

BELO HORIZONTE, Brazil and SINGAPORE, Sept. 08, 2026 (GLOBE NEWSWIRE) -- Magnes Rare Earths (“Magnes”), a wholly owned subsidiary of Verde AgriTech Ltd. (TSX: NPK | OTCQX: VNPKF) (“Verde” or the “Company”), is pleased to report additional head-grade rare earth assay results from 146 drill holes totalling 1,299 assayed metres at the Minas Americas Global Alliance project (“Minas Americas” or the “Project”) in Minas Gerais, Brazil. The results comprise 139 completed auger holes totalling 1,027 m and the first seven completed or reconciled reverse-circulation (“RC”) holes comprising 272 assayed metres.

The new auger results are led by MAV_AD_0375, the highest-grade full-hole, from-surface interval reported at Minas Americas to date. The hole returned 15 m from surface to the end of hole (“EOH”) averaging 11,480 parts per million (“ppm”) total rare earth oxides (“TREO”), or 1.15% TREO, and 2,830 ppm magnetic rare earth oxides (“MREO”), comprising 2,752 ppm NdPr oxides and 79 ppm DyTb oxides. MREO represents 24.7% of TREO in the interval. Within the continuous 15 m profile, one metre from 12 m to 13 m returned 19,952 ppm TREO (2.00% TREO) and 5,909 ppm MREO.

The first RC results add a new vertical dimension to the drill-derived dataset. Five of the seven holes returned 7–29 m intervals averaging 0.53%–0.77% TREO to EOH, including four intervals at least 15 m long. MAV_RC_0012 returned 29 m from 31 m to the 60 m EOH at 5,751 ppm TREO (0.58% TREO) and 1,346 ppm MREO. Two separate five-metre intervals at depths between 38 m and 51 m exceeded 0.90% TREO and 0.23% MREO: MAV_RC_0012 returned 5 m from 38 m to 43 m at 9,106 ppm TREO, 2,315 ppm MREO and 566 ppm heavy rare-earth oxides (“HREO”), including 100 ppm DyTb oxides; MAV_RC_0005 returned 5 m from 46 m to 51 m at 9,191 ppm TREO, 2,429 ppm MREO and 430 ppm HREO, including 91 ppm DyTb oxides.

“Today’s results give us the strongest combination yet of near-surface grade, depth and magnet rare-earth content at Minas Americas,” said Cristiano Veloso, Founder and Chief Executive Officer of Magnes Rare Earths. “Fifteen metres at 1.15% TREO from surface establishes a new full-hole benchmark for the Project. Just as importantly, our first RC drilling has identified multiple broad, higher-grade intervals far below the reach of the auger program. Five of seven RC holes carry intervals grading between 0.53% and 0.77% TREO to their current depth limits, including 29 metres at 0.58% TREO to the bottom of a 60-metre hole. The heavy rare-earth component is also emerging in both shallow and deeper drilling, led by 5 metres at 566 ppm HREO—including 100 ppm dysprosium-plus-terbium oxides—in MAV_RC_0012.”

“Rare-earth projects will ultimately be judged on recoverable magnet products, not head grade alone. The next value-defining work is to test beneath these current depth limits, connect the expanded dataset in three dimensions and determine recoveries from representative drill composites. With 443 holes and more than 3.6 kilometres of reported assays, Minas Americas now has a substantially stronger platform for that work and for the planned maiden mineral resource estimate.”

Prior to this release, Magnes had reported assay results from 297 auger holes totalling 2,329.8 m at Minas Americas. This release adds 139 auger holes totalling 1,027 m and seven RC holes comprising 272 assayed metres, for 146 holes and 1,299 m. Cumulative reported assay drilling now totals 443 holes and 3,628.8 m.

Highlights

• Highest-grade full-hole result reported to date: MAV_AD_0375 returned 15 m from surface to EOH at 11,480 ppm TREO (1.15% TREO) and 2,830 ppm MREO, comprising 2,752 ppm NdPr and 79 ppm DyTb. MREO represents 24.7% of TREO in the interval, which includes one metre at 19,952 ppm TREO (2.00% TREO) and 5,909 ppm MREO.
• Broad interval at depth: MAV_RC_0012 returned 29 m from 31 m to the 60 m EOH at 5,751 ppm TREO (0.58% TREO) and 1,346 ppm MREO, including 14 m at 6,636 ppm TREO and 1,561 ppm MREO and 5 m at 9,106 ppm TREO and 2,315 ppm MREO. No composite crosses the documented 30–31 m no-recovery interval.
• Five of seven RC holes carry elevated grades to EOH: five holes returned 7–29 m intervals averaging 5,258–7,744 ppm TREO (0.53%–0.77% TREO) to their current depth limits, including four intervals at least 15 m long. Mineralization below the current EOH depths remains untested.
• Two deep intervals above 0.90% TREO and 0.23% MREO: MAV_RC_0005 returned 5 m from 46 m to 51 m at 9,191 ppm TREO (0.92% TREO) and 2,429 ppm MREO, while MAV_RC_0012 returned 5 m from 38 m to 43 m at 9,106 ppm TREO (0.91% TREO) and 2,315 ppm MREO.
• Broad mineralization from surface: MAV_RC_0007 intersected 27 m from surface to EOH at 5,258 ppm TREO (0.53% TREO) and 1,207 ppm MREO, including 10 m from 7 m to 17 m at 8,215 ppm TREO (0.82% TREO) and 1,969 ppm MREO.
• High auger grades repeat across multiple holes: four separate auger holes returned complete, from-surface averages of 0.86%–1.15% TREO over 8–15 m. Across the full auger package, 35 of 139 holes—one in four—averaged at least 4,000 ppm TREO (0.40% TREO) over a combined 300 m.
• Heavy rare-earth oxides accompany high-grade TREO: MAV_RC_0012 returned 5 m at 566 ppm HREO, including 100 ppm DyTb, while MAV_RC_0005 returned 5 m at 430 ppm HREO, including 91 ppm DyTb. The 272 assayed RC metres average 145.5 ppm HREO, including 29.8 ppm DyTb.
• Elevated grades at current depth limits: six of the seven RC holes returned more than 4,000 ppm TREO in their final assayed metre, identifying clear targets for deeper follow-up drilling.

• Cumulative scale: reported assay drilling at Minas Americas now stands at 443 holes and 3,628.8 m.
  

The 4,000 ppm and 7,000 ppm thresholds are used only to summarize assay results and are not mineral-resource cut-off grades or economic thresholds.

Figure 1. TREO head grade by drilled depth for the seven RC holes reported in this release. Each column uses a common 0–60 m vertical scale. Highlighted outlines identify selected length-weighted composites of contiguous, original one-metre samples. The hatched interval in MAV_RC_0012 denotes a documented no-recovery interval. Holes are arranged for comparison and are not shown in spatial relationship.

Why These Results Matter

The first RC assays materially extend the drill-derived dataset below auger depths. In MAV_RC_0012, the 29 m lower interval below the documented no-recovery gap averages 5,751 ppm TREO and 1,346 ppm MREO to 60 m—more than eight times the TREO grade and ten times the MREO grade of the upper 30 assayed metres.

Five of seven RC holes returned elevated intervals to EOH and six ended above 4,000 ppm TREO in their final assayed metre, creating specific deeper-drilling targets. Comparable grades below current EOH depths and continuity between holes remain untested.

Why the Magnet-Rare-Earth Basket Matters

Neodymium, praseodymium, dysprosium and terbium—the four oxides in MREO—are used in high-performance permanent magnets. The International Energy Agency’s 2026 Rare Earth Elements report estimates that permanent magnets account for approximately 95% of global rare-earth consumption by value and that demand for these four elements is set to grow by a further one-third by 2030 under current policy settings.

Minas Americas assays quantify all four magnet oxides. Across 272 assayed RC metres, MREO averages 828 ppm, or 22.8% of TREO, led by 798 ppm NdPr with 29.8 ppm DyTb. MAV_AD_0375 averages 2,830 ppm MREO (0.28%); the two deep five-metre RC intervals average 2,315 ppm and 2,429 ppm MREO.

Earlier leach tests indicated ionic adsorption behaviour. These results are head grades and do not establish desorbability or recovery. Representative drill-based metallurgical testwork is the next major test of how the magnet-oxide grades may translate into a recoverable product.

Source: International Energy Agency, Rare Earth Elements (2026), https://www.iea.org/reports/rare-earth-elements.

Heavy Rare-Earth Oxides: The Dysprosium–Terbium Magnet Subset

Beyond the strong TREO and MREO results, the assays quantify a distinct heavy rare-earth component. MAV_RC_0012 returned 5 m from 38 m to 43 m averaging 566 ppm HREO, including 100 ppm combined Dy2O3 and Tb4O7, within the interval grading 9,106 ppm TREO and 2,315 ppm MREO. MAV_RC_0005 returned 5 m from 46 m to 51 m at 430 ppm HREO, including 91 ppm DyTb; MAV_RC_0002 returned 7 m from 18 m to EOH at 357 ppm HREO, including 69 ppm DyTb; and MAV_AD_0375 returned 15 m from surface to EOH at 322 ppm HREO, including 79 ppm DyTb.

Heavy rare-earth oxides were present in the final assayed metre of every RC hole, ranging from 118 ppm to 345 ppm HREO and from 26 ppm to 75 ppm DyTb. These end-of-hole assays strengthen the rationale for deeper follow-up, but do not establish grade continuity below the current hole limits. Across all 272 assayed RC metres, HREO averages 145.5 ppm, including 29.8 ppm DyTb.

Figure 2. Heavy rare-earth oxide head grades in selected auger and RC intervals reported in this release. Stacked bars partition HREO into Y2O3, combined Dy2O3 and Tb4O7, and the remaining defined heavy rare-earth oxides. Bar length represents grade, not interval thickness. These are head-grade assays and do not establish recovery, mineralogical deportment, continuity or economic value.

For this release, HREO is defined as the sum of Y2O3, Dy2O3, Tb4O7, Ho2O3, Er2O3, Tm2O3, Yb2O3 and Lu2O3. DyTb is the combined Dy2O3 and Tb4O7 heavy-magnet subset. MREO overlaps HREO through Dy2O3 and Tb4O7 and the two metrics are not additive. Representative metallurgical testing is required to determine recoverability.

Table 1: Selected Auger Hole-Level Results

Hole	From (m)	To (m)	Interval (m)	TREO (ppm)	TREO (%)		MREO (ppm)	MREO/TREO		NdPr (ppm)	DyTb (ppm)
MAV_AD_0375	0	15 EOH	15	11,480	1.15	%	2,830	24.7	%	2,752	79
MAV_AD_0381	0	13 EOH	13	9,380	0.94	%	2,555	27.2	%	2,481	74
MAV_AD_0387	0	14 EOH	14	8,930	0.89	%	2,275	25.5	%	2,208	67
MAV_AD_0566	0	8 EOH	8	8,565	0.86	%	1,999	23.3	%	1,930	68
MAV_AD_0418	0	6 EOH	6	6,627	0.66	%	1,601	24.2	%	1,547	55
MAV_AD_0447	0	14 EOH	14	6,195	0.62	%	1,390	22.4	%	1,343	47

Table 1 presents selected full-hole, from-surface auger composites. Complete hole-level results and drill-collar information for the holes reported in this release will be set out in the accompanying Minas Americas Drill-Hole and Assay Results Annex — 8 Sep 2026.

Table 2: Selected RC Results

Hole	From (m)	To (m)	Interval (m)	TREO (ppm)	TREO (%)		MREO (ppm)	MREO/TREO		NdPr (ppm)	DyTb (ppm)
MAV_RC_0012*	0	30	30	705	0.07	%	131	18.6	%	125	6
MAV_RC_0012*	31	60 EOH	29	5,751	0.58	%	1,346	23.4	%	1,294	52
including	31	45	14	6,636	0.66	%	1,561	23.5	%	1,495	66
including	38	43	5	9,106	0.91	%	2,315	25.4	%	2,215	100
MAV_RC_0007	0	27 EOH	27	5,258	0.53	%	1,207	23.0	%	1,164	43
including	7	17	10	8,215	0.82	%	1,969	24.0	%	1,906	63
MAV_RC_0009	0	40 EOH	40	4,295	0.43	%	1,004	23.4	%	970	33
including	25	40 EOH	15	7,744	0.77	%	1,911	24.7	%	1,851	60
MAV_RC_0005	0	56 EOH	56	2,971	0.30	%	655	22.0	%	630	24
including	41	56 EOH	15	6,750	0.68	%	1,675	24.8	%	1,616	59
including	46	51	5	9,191	0.92	%	2,429	26.4	%	2,337	91
MAV_RC_0008	0	50 EOH	50	3,502	0.35	%	810	23.1	%	782	29
including	25	40	15	6,018	0.60	%	1,450	24.1	%	1,400	50
MAV_RC_0002	0	25 EOH	25	3,763	0.38	%	847	22.5	%	817	30
including	18	25 EOH	7	7,652	0.77	%	1,898	24.8	%	1,829	69
MAV_RC_0006	0	15 EOH	15	3,255	0.33	%	741	22.8	%	716	25

Note: The asterisk identifies MAV_RC_0012, for which assay coverage comprises 0–30 m and 31–60 m. The 30–31 m interval was not recovered and therefore has no assay result. No composite crosses the gap. Table 2 intervals are length-weighted averages of contiguous, original one-metre samples. Complete results and drill-collar information for the holes reported in this release will be set out in the accompanying Minas Americas Drill-Hole and Assay Results Annex — 8 Sep 2026.

Reported Drilling Recap

Reporting period	Drill type	Holes	Assayed metres
Reported before this release	Auger	297	2,329.8
Present release	Auger	139	1,027.0
Present release	RC	7	272.0
Present release subtotal	Auger + RC	146	1,299.0
Cumulative reported total	Auger + RC	443	3,628.8

The previously reported total comprises 44 initial auger holes totalling 473.8 m and 253 additional auger holes totalling 1,856 m announced on 16 June 2026. Each hole is counted once. The RC metre total counts original drill intervals for which assays are reported; the documented one-metre no-recovery interval in MAV_RC_0012 is excluded.

Minas Americas Project Background

Minas Americas is an exploration-stage clay-hosted rare earth project across 13 mineral rights in Minas Gerais, Brazil. Surface and trench sampling returned up to 8,930 ppm TREO and 2,182 ppm MREO before drilling commenced. Initial leach tests indicated ionic adsorption behaviour, with up to 667 mg/kg desorbable rare earth oxides (“DREO”) and up to 278 mg/kg MREO in primary leach solution.

Initial drilling results were announced on 2 December 2025, followed on 26 January 2026 by a 13 m interval at 0.83% TREO, including 8 m at 1.01% TREO. On 16 June 2026, Magnes reported assays from 253 additional holes and 1,856 m, led by MAV_AD_0200, which returned 10 m from surface at 9,736 ppm TREO and 2,407 ppm MREO. That release brought the previously reported total to 297 holes and 2,329.8 m.

From Shallow Discovery to Three-Dimensional Definition

With 443 holes and 3,628.8 m of reported assays, Minas Americas is progressing from a shallow auger discovery toward three-dimensional geological definition. The first RC results materially increase the vertical range of the dataset, identify priority targets below current EOH depths and provide drill-derived material to support representative metallurgical composite selection.

Minas Americas has no mineral resource, mineral reserve, project economics or development decision. The next value-defining questions are whether deeper and follow-up drilling reproduces the elevated intervals, how the shallow and deeper profiles connect in the geological model and how representative mineralized material performs in controlled metallurgical testwork.

Next Milestones

Planned workstreams include:

• integrating the expanded auger and RC assay dataset into the three-dimensional geological model;
• priority follow-up RC drilling around the principal broad, higher-grade intervals and beneath selected holes with elevated TREO in their final assayed metre;
• selecting representative drill-based composites and advancing metallurgical testwork;
• continued resource-definition and step-out drilling, with additional holes reported only after assay coverage has been completed or reconciled and the applicable QA/QC review has been closed;
• publication of maiden S-K 1300 and NI 43-101 mineral resource estimates before the end of 2026, subject to drilling results, assay turnaround, technical work and regulatory review;
• publication of a preliminary economic assessment in H1 2027, subject to completion of the required resource, metallurgical, engineering and economic work; and
  continued strategic development and capital-markets preparation for Magnes, including the intended U.S. listing process if conditions and approvals support it.

Technical Notes

Assays are reported as head grades in ppm. 10,000 ppm equals 1.0%. TREO means the sum of the oxides of lanthanum through lutetium plus yttrium. MREO means the sum of Nd2O3, Pr6O11, Dy2O3 and Tb4O7. HREO means the sum of Y2O3, Dy2O3, Tb4O7, Ho2O3, Er2O3, Tm2O3, Yb2O3 and Lu2O3. NdPr means the sum of Nd2O3 and Pr6O11. DyTb means the sum of Dy2O3 and Tb4O7. MREO and HREO overlap through Dy2O3 and Tb4O7 and are not additive. DREO, where prior leach testing is referenced, means desorbable rare earth oxides.

Selected composites are length-weighted averages of contiguous, original one-metre drill-interval samples. Summary calculations use head-grade assay values. For calculation purposes, values reported below a detection limit were assigned one-half of that limit, while values reported above an upper reporting limit were assigned the stated limit. Rounding may result in minor differences.

Auger and RC samples represent one-metre drilled intervals. Drill-interval calculations exclude QA/QC, duplicate and other control samples.

All reported auger and RC holes were drilled vertically, with a dip of −90°; azimuth is not applicable to vertical holes. Reported intervals are drilled lengths; true thickness has not been determined. Collar coordinates, elevations and EOH depths will be set out in the accompanying Minas Americas Drill-Hole and Assay Results Annex — 8 Sep 2026. Spatial coordinates are referenced to the SIRGAS 2000 geodetic datum and projected using the Universal Transverse Mercator (UTM) coordinate system, Zone 23S.

Six of the seven RC holes have continuous assay coverage from surface to EOH. MAV_RC_0012 contains a documented no-recovery interval from 30 m to 31 m. No composite crosses that interval. Every included hole record has been reconciled to its recorded EOH. Incomplete holes and holes subject to an unresolved QA/QC hold are excluded from this release.

QA/QC

Auger and RC samples were submitted to SGS Geosol Laboratórios S.A. in Vespasiano, Minas Gerais, Brazil, a third-party commercial laboratory external to Magnes and Verde. Analyses used lithium-borate fusion followed by ICP-MS/OES determination for rare earth and oxide analysis. Magnes’ quality assurance and quality control procedures include the regular insertion of blanks, certified reference materials and field duplicates into the sample stream. Laboratory repeat and duplicate analyses were performed by SGS.

The Qualified Person verified the sampling, analytical and test data supporting this disclosure by reconciling drill-hole, interval, sample and batch identifiers between the field database and laboratory records; comparing analytical values, methods, units and qualifiers with the laboratory records; independently recalculating selected oxide totals and length-weighted composites; and reviewing the applicable blanks, certified reference materials, field duplicates and laboratory controls. No material discrepancies or limitations were identified during this verification. Based on the procedures performed and results obtained, the Qualified Person considers the underlying data reliable and suitable for disclosure of exploration results. QA/QC and control samples are excluded from public assay-result tables unless specifically discussed for QA/QC purposes.

Qualified Person

The scientific and technical information contained in this news release has been reviewed and approved by Carlos Leite, MAusIMM CP(Geo) (No. 1006695), an independent consultant to Magnes and a Qualified Person as defined by National Instrument 43-101 — Standards of Disclosure for Mineral Projects.

About Magnes Rare Earths

Magnes Rare Earths was created as a wholly owned subsidiary of Verde AgriTech Ltd. to hold and advance the Minas Americas Global Alliance Project in Minas Gerais, Brazil. Magnes is intended to serve as Verde’s dedicated rare earths platform and, subject to market conditions and applicable approvals, pursue a future listing on a U.S. stock exchange.

About Verde AgriTech

Verde AgriTech is a Brazil-focused specialty fertilizer company listed on the TSX and OTCQX. Through its wholly owned subsidiary Magnes Rare Earths, the Company also retains exposure to the Minas Americas Global Alliance rare earth project in Minas Gerais, Brazil.

For Additional Information

Website: magnesrareearths.com

Investor relations: magnesrareearths.com/contact?topic=investor#enquiry

Register for Magnes investor updates: magnesrareearths.com/investors#updates

Forward-Looking Statements

This news release contains forward-looking information and forward-looking statements within the meaning of applicable securities laws. These include statements regarding the significance and interpretation of exploration results; the potential continuity, geometry, extent and depth of mineralization; whether mineralization continues below current EOH depths; geological modelling and target ranking; follow-up, step-out and resource-definition drilling; receipt, validation and reporting of additional assays; representative composite selection, metallurgical testwork and technical studies; the potential timing and publication of S-K 1300 and NI 43-101 mineral resource estimates and a preliminary economic assessment; the potential pursuit of a U.S. stock exchange listing for Magnes; and Magnes’ and Verde’s plans, objectives and expectations.

Forward-looking statements are based on management’s current expectations, estimates, interpretations and assumptions, including the accuracy and representativeness of sampling and assay data; the potential continuity of mineralization; the availability of drilling equipment, laboratories, qualified personnel and financing; timely receipt of required permits and approvals; and technical work progressing as presently anticipated. Forward-looking statements involve risks and uncertainties that may cause actual results or events to differ materially, including exploration-stage risk, assay and metallurgical variability, geological interpretation and continuity risk, sampling and QA/QC risk, mineral-right and permitting risk, commodity-market and financing risk, regulatory and exchange approvals, market conditions and other risks described in the Company’s continuous-disclosure filings.

No mineral resource estimate, mineral reserve estimate, production guidance, project-level economic assessment, recovery estimate or development decision is provided in this release. There is no certainty that further exploration will delineate mineral resources or mineral reserves, that metallurgical testwork will demonstrate economic recoveries, or that a development decision will be made. Readers should not place undue reliance on forward-looking statements. Forward-looking statements speak only as of the date of this news release. Magnes and Verde do not undertake to update them except as required by applicable securities laws.

Photos accompanying this announcement are available at:

https://www.globenewswire.com/NewsRoom/AttachmentNg/6f96168f-6e64-4292-b3db-175424ed102d

https://www.globenewswire.com/NewsRoom/AttachmentNg/50cfb4de-fc58-4894-8ccb-2dc40ac6a0b3