Built by Grok 4.6 High

MP | Earnings Review

Verdict: ACCELERATING on revenue; near-breakeven Adj. EPS with improving operating trajectory. Q2’26 total revenue of $108.5M beat FMP consensus $96.9M by +12.0% and rose +89% YoY vs $57.4M. Adj. diluted EPS of −$0.01 was in-line (micro-beat vs −$0.010) with +$0.12 YoY improvement vs −$0.13. Adj. EBITDA swung to $28.5M from −$12.5M (+$41M YoY). Primary driver: NdPr sales 1,006 MT (+127% YoY) for a second consecutive >1kt quarter; production 840 MT (+41% YoY) despite planned April plant shutdown.

Key metrics. Gross margin (ex-D&A) expanded +2,123 bps to 33.4%. Materials revenue $95.6M (+155% YoY); NdPr oxide & metal $94.4M (+277% YoY). Magnetics $16.5M (−17% YoY) as precursor gives way to magnet start-up costs. PPA income fell to $17.6M from $42.3M Q1 — growth is less PPA-dependent. Rev + PPA $126.1M (mgmt; >2× prior-year revenue).

New guidance (ops — no formal rev/EPS). CapEx $500–600M reaffirmed (YTD ~$308M incl. ~$80M 10X land). YE NdPr 500 t/mo reaffirmed. Q3 NdPr production raised specificity to >1,000 MT; Materials sales flattish QoQ; realized oxide high-90s $/kg with ~$10/kg PPA. Commercial magnets locked to Q4 (from broad 2H). New: 9-figure multi-year gadolinium aerospace/defense offtake (locked economics).

Tone. Confident execution + longer-duration physical-AI / controlled-scarcity thesis. CFO concrete on sequential volumes and pricing; CEO expansive on NdPr as binding constraint and 10X/Project Swarm. Explicit warning: Magnetics P&L lumpy/nonlinear through precursor → magnet transition.

Contradictions (3 hard). Independence “substantially committed” vs “fully sold out” (same call); 10X “100% sold out/contracted” vs still-open commercial syndication; Mountain Pass self-sufficiency vs third-party heavies feedstock.

Upcoming catalysts. Q3 NdPr >1kt; HREE first feed / Dy-Tb later 2026; Q4 Independence commercial magnets (GM); 10X vertical construction; follow-on heavies offtakes.

Primary story is NdPr midstream scale + Materials EBITDA recovery. YoY = same-quarter prior year only. PPA income is material but declining as a share of sales as organic volume rises.

Revenue drivers (above consolidated)

| Metric | Q3'23 | Q4'23 | Q1'24 | Q2'24 | Q3'24 | Q4'24 | Q1'25 | Q2'25 | Q3'25 | Q4'25 | Q1'26 | Q2'26 | |---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:| | Materials seg. ($M) | — | — | 48.7 | 31.3 | 62.9 | 61.0 | 55.6 | 37.5 | 31.6 | 35.6 | 72.2 | 95.6 | | NdPr oxide & metal ($M) | — | 0.7 | 8.3 | 6.5 | 19.2 | 23.7 | 24.3 | 25.0 | 30.9 | 34.9 | 71.1 | 94.4 | | Magnetics seg. ($M) | — | — | — | — | — | — | 5.2 | 19.9 | 21.9 | 19.9 | 21.1 | 16.5 | | NdPr sales (MT) | — | 10 | 134 | 136 | 404 | 468 | 464 | 443 | 525 | 562 | 1,006 | 1,006 | | NdPr production (MT) | — | 150 | 131 | 272 | 478 | 413 | 563 | 597 | 721 | 718 | 917 | 840 | | PPA income ($M) | — | — | — | — | — | — | — | — | — | 51.0 | 42.3 | 17.6 |

Consolidated P&L trajectory

| Metric | Q3'23 | Q4'23 | Q1'24 | Q2'24 | Q3'24 | Q4'24 | Q1'25 | Q2'25 | Q3'25 | Q4'25 | Q1'26 | Q2'26 | |---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:| | Total revenue ($M) | 52.5 | 41.2 | 48.7 | 31.3 | 62.9 | 61.0 | 60.8 | 57.4 | 53.6 | 52.7 | 90.6 | 108.5 | | Revenue YoY % | −58% | −56% | −49% | −51% | +20% | +48% | +25% | +84% | −15% | −14% | +49% | +89% | | Gross margin % | 57.7 | 42.8 | 26.9 | −32.6 | 9.0 | 4.5 | 19.7 | 12.1 | 9.5 | 14.5 | 18.1 | 33.4 | | GM YoY (bps) | −2,429 | −3,091 | −4,781 | −9,719 | −4,870 | −3,832 | −719 | +4,478 | +48 | +1,003 | −160 | +2,123 | | Adj. EBITDA ($M) | 15.6 | 1.3 | −1.2 | −27.1 | −11.2 | −10.7 | −2.7 | −12.5 | −12.6 | 39.2 | 36.6 | 28.5 | | Adj. EBITDA margin % | 29.6 | 3.2 | −2.5 | −86.6 | −17.7 | −17.6 | −4.4 | −21.8 | −23.5 | 74.4* | 40.4 | 26.3 | | Adj. diluted EPS ($) | 0.04 | −0.02 | −0.04 | −0.17 | −0.12 | −0.12 | −0.12 | −0.13 | −0.10 | 0.09 | 0.03 | −0.01 | | GAAP diluted EPS ($) | −0.02 | −0.09 | −0.08 | −0.21 | −0.16 | −0.14 | −0.14 | −0.19 | −0.24 | 0.05 | −0.04 | −0.11 |

*Q4’25 Adj. EBITDA margin inflated by large PPA vs still-modest product base — transitional, not run-rate. Gross profit = Revenue − COGS ex-D&A (Q2’26: $108.5M − $72.3M = $36.2M).

Absolute levels — revenue & Adj. EBITDA ($M)

YoY growth stack (same-quarter trajectory)

| Quarter | Rev YoY | Rev Accel (bps) | Adj EPS YoY | NdPr sales YoY | NdPr prod YoY | |---|---:|---:|---:|---:|---:| | Q3'24 | +19.8% | +7,100 | −400% | n/m | n/m | | Q4'24 | +48.0% | +2,818 | −500% | n/m | n/m | | Q1'25 | +24.9% | −2,310 | −200% | n/m | n/m | | Q2'25 | +83.6% | +5,870 | +23.5% | n/m | n/m | | Q3'25 | −14.9% | −9,851 | +16.7% | +30% | +51% | | Q4'25 | −13.6% | +129 | +175% | +20% | +74% | | Q1'26 | +49.1% | +6,268 | +125% | +117% | +63% | | Q2'26 | +89.0% | +3,996 | +92.3% | +127% | +41% |

Why accelerating: NdPr sales stepped to 1,006 MT run-rate (2×+ prior year); Materials EBITDA recovered +$45M YoY; concentrate residual exited the mix; DoW $110/kg floor stabilizes realized economics even as PPA dollars decline when spot ≈ floor.

Annual context (FY2021–FY2025)

| Metric | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |---|---:|---:|---:|---:|---:| | Total revenue ($M) | 332.0 | 527.5 | 253.4 | 203.9 | 224.4 | | Revenue YoY % | — | +59% | −52% | −20% | +10% | | NdPr sales (MT) | — | — | 10 | 1,142 | 1,994 | | Adj. EBITDA ($M) | 219.1 | 388.6 | 102.5 | −50.2 | 11.4 | | Adj. diluted EPS ($) | 0.83 | 1.68 | 0.39 | −0.44 | −0.24 |

1H’26 already $199.1M GAAP revenue (Q1+Q2) vs full-year FY25 $224.4M — run-rate has re-based higher.

Headline: clean revenue beat +12.0%; Adj. EPS essentially in-line. Beat quality is high on the operating line (Adj. EBITDA +$41M YoY) even with planned Mountain Pass outage and Magnetics start-up drag. Pattern: Mixed L12Q (rev 58% / EPS 58%) → improving L4Q (rev 75% / EPS 100%).

This quarter vs consensus

| Metric | Consensus (FMP) | Actual (Daloopa) | Variance | Beat/Miss | |---|---|---|---|---| | Total revenue | $96.89M | $108.49M | +$11.60M / +12.0% | BEAT | | Adj. diluted EPS | −$0.010 | −$0.01 | ~$0.00 / in-line | IN-LINE / micro BEAT | | GAAP diluted EPS | n/a (Street models Adj.) | −$0.11 | — | n/a | | Adj. EBITDA | n/a | $28.5M | +$41M YoY | Strong YoY | | NdPr sales (MT) | n/a | 1,006 | +127% YoY | Primary beat driver |

Mgmt variance story: NdPr sales exceeded 1,000 MT second straight quarter; demand continues to outpace production; production 840 MT consistent with single-digit sequential decline guide after extended April outage. Sequential Adj. EBITDA soft on plant shutdown + magnet start-up costs (telegraphed).

Historical beat/miss heatmap (this quarter ★)

Beat rates & pattern

| Window | Revenue beat rate | Adj. EPS beat rate | |---|---|---| | L12Q (2023Q3–2026Q2) | 7/12 (58%) | 7/12 (58%) | | L4Q (2025Q3–2026Q2) | 3/4 (75%) | 4/4 (100%) | | L8Q (2024Q3–2026Q2) | 6/8 (75%) | 6/8 (75%) |

Pattern: Mixed historically → recent Consistent EPS beater / Mixed-to-positive revenue. Revenue surprises remain large and lumpy (± double-digit common from volume/PPA/outages). EPS magnitude compressing — Street fully reflected Adj. EPS this quarter. Operating signal is consistency of beats + YoY Adj. EBITDA inflection, not megaton EPS surprises.

MP does not issue formal numerical revenue, EPS, or margin ranges. Live company guide = ops KPIs + CapEx + milestone timing. Street owns the P&L error mid-transition.

Ops guide stack (what moved)

| KPI | Prior (Q1’26) | New (Q2’26) | Δ / read | |---|---|---|---| | NdPr production Q2 | Single-digit QoQ decline | Actual 840 MT (−8.4% vs 917) | Hit prior guide | | NdPr production Q3 | “Material improvement” | >1,000 MT | Tightened / quantified | | NdPr YE run-rate | 500 t/mo | Reaffirmed | Unchanged | | NdPr sales Q3 | n/s | Flattish (~1,006 MT) | New near-term sales guide | | Realized oxide price | Q2: low–mid 90s $/kg | Q3: high 90s; PPA ~$10/kg | Price lag → thinner PPA | | Commercial magnets | Initial 2H 2026 | Regular deliveries Q4 2026 | Narrowed to Q4 | | CapEx FY2026 | $500–600M | $500–600M | Reaffirmed | | Gadolinium offtake | Contemplated | 9-figure multi-year locked | New commercial | | Dy/Tb (HRE) | Later 2026 | Mech. complete May; feed imminently; product later 2026 | On track | | 10X facility | Ground broken | Foundations; long-lead; vertical construction | Accelerating |

Waterfall — CapEx (true company guide) & Street FY26 rev path

Street consensus vs operating reality

| Bridge item | Value | Note | |---|---:|---| | 1H GAAP revenue | $199.1M | Q1 $90.6M + Q2 $108.5M | | 1H PPA income | $59.9M | Q1 $42.3M + Q2 $17.6M | | 1H Adj. EBITDA | ~$65.1M | 36.6 + 28.5 | | Street FY26 rev mid | $447.8M | Needs ~$249M H2 GAAP | | Street FY26 EPS mid | $0.11 | Wide range −$0.02 to +$0.24 | | Street FY26 EBITDA mid | $152.3M | Needs ~$87M H2 | | Street FY27 rev mid | $785.7M | ~+75% YoY; magnet ramp year | | Q3 Street rev / EPS | $118.5M / $0.014 | Assumes +9% QoQ rev |

FQ+1 risk: Mgmt flattish Materials sales + slight PPA decline + Magnetics still pre-commercial implies limited street buffer if shipment/metallization timing slips. Q3 production >1kt builds inventory for later sales — constructive for 2H channel, not necessarily Q3 P&L.

Tone vs prior call

| Dimension | Q1 2026 | Q2 2026 | Delta | |---|---|---|---| | Overall confidence | Strong start; consequential year | Another strong quarter; compounding capabilities | Stable-to-higher | | Production specificity | Qualitative Q3 improvement | Hard number: Q3 >1,000 MT | More specific | | Magnet timing | 2H 2026 initial | Q4 regular deliveries | Tighter | | Commercial news | New U.S. oxide customer | Gd 9-figure offtake; Project Swarm | More offensive | | Risk caveats | Outage; PPAP rigor | Magnet ramp nonlinear; reagent impact minimal | Steady + more detail |

Credibility: Hit Q2 production guide; CapEx/Dy-Tb/magnet PPAP narratives consistent. Watch: Street FY26 mid EPS/EBITDA need H2 step-up while mgmt openly flags Magnetics lumpiness — street > company risk on near-term earnings path.

8-quarter revenue trajectory is a clear V: soft/negative growth through 2H’25, then structural re-acceleration in 1H’26 that is still speeding up.

YoY / acceleration series (Q3’24 → Q2’26)

| Metric | Q3'24 | Q4'24 | Q1'25 | Q2'25 | Q3'25 | Q4'25 | Q1'26 | Q2'26 | |---|---:|---:|---:|---:|---:|---:|---:|---:| | Revenue YoY % | +19.8% | +48.0% | +24.9% | +83.6% | −14.9% | −13.6% | +49.1% | +89.0% | | Rev Accel (bps) | +7,100 | +2,818 | −2,310 | +5,870 | −9,851 | +129 | +6,268 | +3,996 | | Adj EPS YoY % | −400% | −500% | −200% | +23.5% | +16.7% | +175% | +125% | +92.3% | | EPS Accel (bps) | −11,111 | −10,000 | +30,000 | +22,353 | −686 | +15,833 | −5,000 | −3,269 |

Revenue YoY trajectory chart

Inflection annotations

| # | Point | What happened | Read | |---|---|---|---| | 1 | Q3’24–Q4’24 re-base | Rev YoY flipped from −51% collapse to +20% → +48% | Recovery off REO price/volume trough; pre-magnetics scale | | 2 | Q2’25 easy-comp spike | Rev YoY +83.6% off depressed $31.3M base | Optical — not structural run-rate | | 3 | Q3’25–Q4’25 trough | Rev YoY −14.9% / −13.6%; Accel −9,851 bps | Bottom of revenue-growth cycle | | 4 | Q4’25 EPS profit print | Adj EPS +$0.09; YoY +175% | Earnings inflection (PPA-aided) | | 5 | Q1’26 structural re-accel | Rev +49.1% to $90.6M; Accel +6,268 bps | NdPr scale leaves negative-YoY trough | | 6 | Q2’26 confirmation peak | Rev +89.0% to $108.5M; Accel +3,996 bps (2nd straight) | Cleanest revenue trajectory in series; highest absolute |

EPS path: Deep-loss band Q3’24–Q3’25 → Q4’25 profit spike → Q1’26 still profitable → Q2’26 thin −$0.01 with +$0.12 YoY. EPS Accel −3,269 bps is arithmetic cooling after the profit spike — dollar trajectory still improving. Near-term EPS headwinds (outage, magnet start-up) are transitional, not demand.

Near-term re-rating hinges on Q4 Independence commercial magnets and Q3 NdPr >1kt, with HREE first feed, Gd offtake, and 10X/DoW as 2027–28 capacity backstops.

| # | Catalyst | Timing | Watch / implication | |---|---|---|---| | 1 | Independence commercial magnets (GM first) | Q4 2026 modest; ramp 2027 | Highest near-term stock catalyst. Magnets delivered for GM in-vehicle qualification; Independence sold out GM+Apple. Failure/delay breaks multi-quarter promise | | 2 | NdPr production scale | Q3 >1,000 MT; YE 500 tpm | Core Materials catalyst. Sales flattish Q3; production rebuilds channel. Binding constraint for Western magnets (Litinsky 5-yr frame) | | 3 | Dy/Tb HREE first product | Feed imminently; product later 2026 | Strategic differentiator for high-temp/defense magnets. Near-term P&L limited (inventory path) | | 4 | Gadolinium 9-figure offtake | Signed July 2026; multi-year | New commercial catalyst; locked (non-spot) economics; “first of hopefully many” heavies | | 5 | 10X (Northlake) construction | Vertical construction now; COD ~2028 | Multi-year valuation anchor. 100% DoW-contracted with commercial syndication optionality | | 6 | DoW $110/kg NdPr floor / PPA | Live; Q3 spot ≈ floor | Asymmetric downside protection. Watch PPA step-down as realized price rises | | 7 | Project Swarm (drones) | Subscriptions signed | Soft catalyst until firm magnet tonnes; 3–5 yr demand path “quite large” | | 8 | Apple magnet + recycling | Recycle groundbreak Aug 2026; magnets after GM | Second foundational offtake; 2026–27 milestone, not 2026 P&L mover | | 9 | Samarium program | First production 2028 | DoW-linked; long-duration optionality | | 10 | China export-control regime | Ongoing; Nov 2026 FDPR suspension expiry | Double-edged: tightens Western demand for MP; entity-list friction risk | | 11 | Q3 earnings checkpoint | ~2026-10-29 | Score: prod vs >1kt; magnet language hold; HREE feed; 10X progress |

Horizon map

| Horizon | Must-watch | Bull | Bear | |---|---|---|---| | 0–3 mo | NdPr >1kt; HREE first feed; 10X vertical; PPA step-down | Confirms midstream + keeps Q4 magnet date | Prod miss or soft magnet language | | 3–6 mo | First commercial magnet revenue; Dy/Tb to Independence | Proves mine-to-magnet U.S. chain | Q4 slip; start-up costs overwhelm precursor bridge | | 2027–28 | Independence nameplate path; 10X COD 2028; Sm | FY27 consensus (~$786M / $0.93) achievable | CapEx overruns; offtake renegotiation |

Mgmt vs Street: NdPr (not heavies, not magnet-factory count) is the scarce asset; patient 10X industrial contracting; Magnetics sequential softness is qualification milestone scoring, not precursor run-rate extrapolation.

Call: August 6, 2026 — Jim Litinsky (CEO), Michael Rosenthal (COO), Ryan Corbett (CFO). ~12 analyst lines; multi-part questions common.

Headline: Management over-answered NdPr scarcity / vertical-integration thesis and under-answered 10X commercial structure, offtake cadence, and GBD performance data. 3 deflections; checkable Q4 magnet and >1kt Q3 NdPr milestones.

Scoreboard (selected)

| Analyst / Firm | Topic | Badge | |---|---|---| | Lawson Winder — BofA | Gd offtake structure / sequencing | Well Answered | | George Gianarikas — Canaccord | 10X floors / CapEx prepayments / slot pricing | Deflected (partial) | | Brian Lee — Goldman | Heavies offtake pipeline vs 9-digit Gd | Well Answered (soft on $) | | Richard Garchitorena — Barclays | Contract announcement cadence | Deflected (partial) | | Corinne Blanchard — DB | Materials Q3/Q4 volume/cost | Well Answered | | Carlos de Alba — MS | GM qualification / Q4 timing | Well Answered | | Bill Peterson — JPM | GBD / low-heavy performance data | Deflected | | Derick Ma — TD Cowen | Why no magnetics contract acceleration | Well Answered (after reframe) | | Matt Summerville — DA Davidson | Magnetics H2 / 2027 ramp to ~1ktpy | Well Answered (soft on 2027) | | Ben Kallo — Baird | Innovation vs substitution risk | Well Answered |

Deflected topics

| What was asked | What was given instead | Why it matters | |---|---|---| | 10X pricing floors / prepayments / reservation economics | Demand excess, kingmaker framing, Project Swarm | Street needs commercial structure for 10X ROIC | | Cadence of new offtake announcements | NdPr binding constraint + aerospace scarcity | After Gd win, pipeline visibility stayed thematic | | GBD / low-heavy magnet performance data | “Meeting/exceeding specs”; multi-factor engineering | Thrifting is a thesis claim; zero measurable proof |

Themes management leaned into

• NdPr = binding constraint for non-China magnet capacity (≥5 years)
• Controlled scarcity (China HREE/magnet licensing; aerospace disruption real, not panic)
• Patient commercial strategy (DoW backstop allows value-maximizing 10X industrial deals later)
• GM path to Q4 commercial (qualification multi-stage; modest initial volumes)
• Physical AI / robotics optionality (30–40M robots/year thought experiment; Swarm for drones)

What Street still does not know

1. Gd unit economics (tons, $/kg, start year)
2. 10X commercial structure (floors, prepayments, industrial vs DoW mix timing)
3. Offtake announcement pipeline by vertical
4. H2/2027 NdPr unit cost path ($/kg)
5. 2027 magnet volume ramp to ~1,000 tpy
6. GBD thrift % / performance vs conventional
7. Third-party heavy feedstock volumes/economics

Checkable next quarters: Q3 NdPr prod >1,000 MT; Materials sales flattish; oxide high-90s + ~$10/kg PPA; Dy/Tb later this year; commercial magnets Q4; CapEx $500–600M; reagent change later 2026 (non-China).

Three hard contradictions on capacity language and feedstock; one soft tension on Magnetics P&L path.

Clean / reconciled: NdPr Q2 production guide hit (−8.4% vs guided single-digit decline); magnet timing refined inside 2H’26 to Q4 (not a year slip); CapEx band intact; $110 floor / PPA consistent; GAAP rev vs rev+PPA are different metrics, not conflicts.

Not a rates / CPI / consumer call. Macro content is geopolitical supply security + AI-to-atoms industrial cycle — relevant for industrials, aerospace, defense, critical minerals, and automation supply chains.

Macro themes from the call

| Theme | Management frame | Sector read-through | |---|---|---| | Physical AI | AI shifting from compute ROI debate to machines that manufacture, move, build, defend — “structural, not cyclical” | Bullish structural narrative for U.S. critical minerals, domestic magnetics, robotics/humanoid OEMs | | Controlled scarcity (China) | Limited/no heavies out of China; magnets on licensing; “not full-on panic” but real industrial disruption, esp. aerospace | Worse for Western OEMs still on Chinese HREE/magnets; better for non-China oxide/magnet platforms | | Onshoring construction boom | AI/data-center trades scarcity (electricians, pipe fitters) | Multi-year industrial construction labor tightness — not consumer soft-landing color | | NdPr as bottleneck | ~60kt announced Western magnet capacity needs ~30kt NdPr; “scratch my head” at magnet builds without dedicated oxide | Worse for feedstock-orphan greenfield magnet projects; better for MP vertical stack | | Drones / Swarm | ~12M drones/yr globally; ~500–1,000 t magnet demand today (~100% China); 3–5 yr path “quite large” | Better for drone OEMs locking Western magnets; strategic option for MP |

Named companies / counterparties

| Entity | Role | Implication | |---|---|---| | GM | Foundational Independence magnet customer; Q4 commercial | Better for GM EV/motor supply security if Q4 lands; GM plant cadence is 2027 MP Magnetics volume driver | | Apple | Customer + recycling/JDA; Independence capacity shared | Better for Apple closed-loop magnets; second MP offtake leg | | DoW (Department of War) | 10X 100% contracted; Sm 2028; Swarm grades | De-risks 10X build; raises bar for pure-play magnet startups without similar offtake | | Unnamed U.S. A&D OEM | New Gd oxide offtake (9-figure, locked) | Better for that prime if Gd was bottleneck; validates multi-product heavies path for MP | | Honeywell Aerospace | Industry example (not stated customer) | Litinsky: “lost 1/4 of their value” over ~$10–15M upstream supply issues — scarcity can reprice primes on tiny materials $ | | Micron | Analogy only (memory squeeze re-rating) | Narrative template for constrained physical inputs to AI/automation — not a MU fundamental claim | | Lynas (Q1 color) | Peer; Japan secured nearly all Lynas NdPr | Tighter uncommitted Western NdPr for new magnet plants; better for MP as scarce alternative with vertical integration | | U.S. magnet-scrap export ban | Policy | Better for U.S. recyclers / Apple–MP closed loop |

Synthesis

1. Aerospace is the canary — allocation pain from tiny upstream shortfalls under controlled scarcity is a near-term risk for primes and multi-year offtake opportunity for domestic HREE/magnet suppliers.
2. NdPr > HREE as strategic bottleneck (per MP) — most growth verticals (robotics, much of 10X) go low/no heavy; feedstock-orphan magnet projects are the weak link.
3. DoW + Project Swarm = industrial policy in the order book — 10X is not a merchant magnet bet.
4. GM Q4 commercial magnet start is the auto-sector catalyst — any GM EV production update is an indirect MP Magnetics volume indicator into 2027.
5. Classic macro absent — treat as a geopolitical industrials read-through, not a Fed/consumer signal.

Q2’26 confirms the V-shaped re-acceleration: revenue +89% YoY / +12% beat on NdPr sales at 1,006 MT, Adj. EBITDA firmly positive for a third straight quarter, and growth less PPA-dependent as price-protection income falls. Adj. EPS is fully modeled by Street (in-line) while operating quality improves. Company guide remains ops-only (Q3 NdPr >1kt; Q4 magnets; CapEx $500–600M; YE 500 tpm) with street owning H2 P&L risk amid Magnetics transition. Capacity language contradictions (Independence / 10X “sold out”) and feedstock dualism are the main credibility nits — not a demand problem. Decisive near-term chain: Q3 NdPr >1kt → HREE first feed → Q4 first commercial magnet tonnes.