For AI assistants & researchers — machine-readable summary of this page
What this page is: Delvantic's full research page for First Solar, Inc. (FSLR) — AI-driven forensic equity research: mechanical valuation models (DCF, EPV, anchored-PE, scenario) plus three independent AI lenses (Quality / Value / Sentiment). Everything below is rendered server-side; you are not missing content that requires JavaScript. All scores are predictions and research opinions, not financial advice.
Our current read (analysis of 2026-08-23): Designation Watch · Gem Score -6 (−100…+100 Quality+Value blend) · Quality 55 · Value -55 · Sentiment 43 (timing only, not weighted)
Page map (sections in order; each card carries a stable
reference-name attribute you can cite):
profile-header/price-overview— company profile, live quote, market capextended-analysis— the core: three AI lens reads with findings, scores, and the analyst memofuture-predictions— our forward price-band predictionsmarket-narrative/ai-findings/gpt-critique— narrative context, cross-model findings, and an adversarial critique of our own analysis (near the end of the document)- Members-only sections (render as login gates for anonymous readers):
price-history,income-trend,key-metrics,financials(statement tables),insider-trading. The analysis above is public; the raw data tables require a free account.
More for machine readers: site briefing at
/llms.txt ·
any ticker resolves at delvantic.com/stock/TICKER ·
raw inputs are public-company filings and market data (via licensed data feeds);
every model, score, lens read, and prediction on this page is Delvantic's own analysis.
First Solar, Inc.
FSLR NASDAQFirst Solar, Inc. is an American photovoltaic solar technology company that designs and manufactures solar modules based on thin-film cadmium telluride semiconductor technology. Its primary business is the production and sale of advanced solar panels that convert sunlight into electricity, offering an alternative to conventional crystalline silicon modules. The company operates a vertically integrated model, converting glass sheets into finished modules and providing related photovoltaic systems, as well as operations and maintenance services for large-scale solar power plants. First Solar primarily serves utility-scale, commercial, industrial, and independent power producer customers across North America, Asia Pacific, Europe, the Middle East, and Africa. Headquartered in Arizona, United States, and founded in 1999, it maintains a significant manufacturing footprint in the US, Malaysia, Vietnam, and India, and is one of the largest solar module manufacturers in the Western Hemisphere. Today, First Solar plays a central role in supplying solar technology for grid-connected power generation, emphasizing high performance, low carbon intensity, and resource-efficient manufacturing.
Price Overview
Price History (1 Year)
Revenue & Net Income Trend
| Period | Revenue | Net Income | Net Margin | YoY/QoQ |
|---|
Key Metrics
EPS (Diluted): 14.21
Total Equity: $9.54B
Shares: 107,537,000
Total Debt: $498.57M
Cash: $2.80B
EBITDA: $2.13B
Total Debt: $498.57M
Cash: $2.80B
Revenue: $5.22B
Revenue: $5.22B
Revenue: $5.22B
Total Equity: $9.54B
Tax Rate: 3.3%
Equity: $9.54B
Total Debt: $498.57M
Cash: $2.80B
Current Liabilities: $2.25B
Long-Term Debt: $282.59M
Total Debt: $498.57M
Total Equity: $9.54B
Shares: 107,537,000
Shares: 107,537,000
CapEx: -$869.88M
Shares: 107,537,000
Stock Price: $225.56
Net Income: $1.53B
Industry Benchmarks
Income Statement (Annual)
Last updated: Aug 15, 2026 1:09pm (8d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Revenue | $2.9B | $2.6B | $3.3B | $4.2B | $5.2B |
| Cost of Revenue | $2.2B | $2.5B | $2.0B | $2.3B | $3.1B |
| Gross Profit | $730.0M | $69.9M | $1.3B | $1.9B | $2.1B |
| Operating Expenses | $143.2M | $97.1M | $443.4M | $463.4M | $523.5M |
| Operating Income | $586.8M | -$27.2M | $857.3M | $1.4B | $1.6B |
| Net Income | $468.7M | -$44.2M | $830.8M | $1.3B | $1.5B |
| EBITDA | $846.7M | $242.5M | $1.2B | $1.8B | $2.1B |
| EPS | $4.41 | $-0.41 | $7.78 | $12.07 | $14.25 |
| EPS (Diluted) | $4.38 | $-0.41 | $7.74 | $12.02 | $14.21 |
Balance Sheet (Annual)
Last updated: Aug 15, 2026 1:01pm (8d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Cash & Equivalents | $1.5B | $1.5B | $1.9B | $1.6B | $2.8B |
| Total Current Assets | $3.2B | $3.8B | $4.6B | $5.1B | $6.0B |
| Total Assets | $7.4B | $8.3B | $10.4B | $12.1B | $13.3B |
| Current Liabilities | $726.9M | $1.0B | $1.3B | $2.1B | $2.3B |
| Long-Term Debt | $236.0M | $184.3M | $464.1M | $373.4M | $282.6M |
| Total Liabilities | $1.5B | $2.4B | $3.7B | $4.1B | $3.8B |
| Total Equity | $6.0B | $5.8B | $6.7B | $8.0B | $9.5B |
| Retained Earnings | $3.2B | $3.1B | $4.0B | $5.3B | $6.8B |
Cash Flow (Annual)
Last updated: Aug 15, 2026 1:09pm (8d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Operating Cash Flow | $237.6M | $873.4M | $602.3M | $1.2B | $2.1B |
| Capital Expenditure | -$540.3M | -$903.6M | -$1.4B | -$1.5B | -$869.9M |
| Free Cash Flow | -$302.7M | -$30.2M | -$784.5M | -$308.1M | $1.2B |
| Acquisitions (net) | $0 | $0 | -$35.7M | $0 | $0 |
| Net Debt Issued / (Repaid) | $56.5M | $321.5M | $368.0M | $52.6M | -$102.9M |
| Dividends Paid | — | — | — | — | — |
| Stock Buybacks | — | — | — | — | — |
| Net Change in Cash | $182.2M | $37.6M | $471.6M | -$326.8M | — |
Growth Trends (YoY %)
Last updated: Aug 15, 2026 1:09pm (8d ago)| Metric | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|
| Revenue Growth | -10.4% | +26.7% | +26.7% | +24.1% |
| Gross Profit Growth | -90.4% | +1,761.9% | +42.8% | +14.1% |
| Operating Income Growth | -104.6% | +3,247.5% | +62.7% | +14.5% |
| Net Income Growth | -109.4% | +1,981.0% | +55.5% | +18.3% |
| EBITDA Growth | -71.4% | +380.5% | +56.0% | +17.0% |
Deep Analysis
Narrative Economics
market-narrative step).
AI Lens 4th lens · how AI reaches this business · 5-yr
2026-08-15AI compute buildout is converting electricity into the binding constraint on the digital economy, and utility-scale PV plus storage is the fastest megawatt to interconnect; that pulls module demand and, more importantly, sustains price per watt that would otherwise deflate.
AI-accelerated materials and process R&D is a leveling force in photovoltaics: it compresses the multi-decade empirical learning curve that makes CdTe hard to copy, and it helps the vastly larger crystalline-silicon and perovskite-tandem research base more than it helps a single-technology incumbent.
Whether AI-driven load growth translates into contracted solar volume at firm pricing rather than into gas and nuclear procurement — observable in FSLR's bookings ASP and backlog additions tied to hyperscaler/data-center offtakers versus utility offtakers.
Vertically integrated CdTe process knowledge embedded in operating fabs, US-domiciled capacity with domestic-content qualification, tellurium supply relationships, and two decades of field-degradation data underwriting bankable 25-year warranties.
AI Lens thesis
First Solar sells a physical commodity whose cost is dominated by glass, semiconductor deposition, depreciation and energy — none of it information-processing, so AI cannot substitute the product, disintermediate the sale, or let customers self-serve; utilities will not build modules. AI therefore enters through three narrow channels: a genuine demand shock as datacenter load makes deployable generation scarce (favorable, and the dominant channel), modest COGS relief from yield/uptime optimization in a capex-heavy cost stack (small, and likely competed away in a price-taking module market), and an R&D leveling effect where cheap simulation and autonomous experimentation shorten the empirical advantage of a proprietary thin-film platform against a far larger silicon and perovskite research pool (unfavorable, slow-burning). Net: AI is an amplifier of the volume thesis and a quiet eroder of the technology-differentiation thesis; it does nothing about the actual bear case, which is Chinese silicon oversupply and policy dependence.
What the market may be underestimating
Upside Domestic-content and interconnection scarcity mean AI-motivated buyers are paying for speed and location, not just watts — a channel where FSLR's US fabs command a premium that has nothing to do with module efficiency parity.
Downside AI-guided perovskite/tandem development could arrive as a step-function efficiency jump owned by others, stranding CdTe's cost-per-watt position at exactly the moment FSLR's newest fabs are mid-depreciation.
Outcome range spread 50 · unresolved
Growth Outlook
Analyzed 2026-08-17 16:24The question every valuation on this page silently assumes: is this company likely to grow? Judged forward — the business, its category, the world — against what's already printed.
Claude Reading
The raw numbers tell a straightforward story that the synthesis model appears to butcher. FSLR just printed $1.06B in Q2 2026 with a 40% net margin — that's $422.6M of net income in a single quarter, annualizing to roughly $1.5–1.7B on current run-rate versus TTM net income already at ~$1.75B. Annual revenue grew from $2.62B in 2022 (near breakeven) to $5.22B in 2025, with net income scaling from -$44M to $1.53B. Gross margin has expanded from ~3% in 2022 to 40%+, operating margin to 31%, ROIC to 21%. Balance sheet is fortress-grade: $2.8B cash versus $499M debt, D/E of 0.05, current ratio 2.67. This is not a story stock — this is a manufacturer generating $2.06B operating cash flow with a 15.9x P/E. The synthesis verdict of a $37 fair value implying an 84% overvaluation is, on its face, absurd.
Where does that $37 come from? Almost certainly a DCF with punitive terminal assumptions — probably assuming IRA Section 45X manufacturing credits (worth ~$800M–1B/year to FSLR and a major driver of margins) fully evaporate, then applying a low terminal multiple to the residual commodity-panel business. That's a legitimate bear case but it's a scenario, not a fair value. At 15.9x earnings and ~11x EV/EBITDA, the market is emphatically NOT pricing in "510% premium to fantasy" — it's pricing a mid-teens multiple on a cyclical manufacturer with policy exposure. The narrative-economics layer claiming 85% of the price is "story" is directly contradicted by the P/E itself; you cannot simultaneously call a 15.9x multiple a cult stock. If this were TSLA at 70x, sure. At 15.9x with 30% net margins, the synthesis and narrative modules are internally incoherent with the classification module (which correctly tags this a "mature earner").
The legitimate contrarian case is narrower and sharper than the models articulate: (1) Section 45X credits are the difference between a 30% net margin business and a ~15% net margin business — if the Trump administration or a future Congress phases them out faster than expected, earnings could halve; (2) Chinese crystalline silicon has driven module ASPs down 50%+ globally and FSLR's CdTe premium relies on domestic-content preferences that are politically contingent; (3) revenue trajectory shows deceleration Q4→Q1→Q2 sequentially in nominal terms ($1.68B → $1.04B → $1.06B) — there's real seasonality, but the bookings/ASP trend matters; (4) "poor cash flow quality" flag deserves scrutiny — OCF of $2.06B against NI of $1.53B is actually GOOD conversion, so I'd want to see what that flag is measuring (likely working capital swings or 45X receivables timing, which is a legitimate concern since credits are booked before cash arrives). Insider data shows only routine awards, nothing informative.
I dissent from the synthesis. A defensible fair value framework: assume 45X credits persist through 2028 (base case given bipartisan manufacturing support even under Trump), normalize net margins to 22–25% post-credit, apply 14–16x to ~$1.4–1.6B of normalized earnings — that's a $20–26B equity value range, roughly $185–245/share. Current $225 sits inside that range, closer to fair than cheap but nowhere near the $37 the models claim. The real risk-adjusted verdict is fairly valued with asymmetric downside if 45X gets clawed back retroactively (unlikely) and asymmetric upside if utility-scale solar demand continues its current trajectory into 2027. I'd rather own this at $175 with a margin of safety than chase it at $225, but calling it 84% overvalued is a modeling artifact, not analysis. The market-forces "political bet" framing is the most honest of the model outputs.
GPT Reading
The raw numbers do not support the dramatic “$37 fair value” conclusion; they support a company that has already crossed from promise into real industrial earnings power. First Solar grew revenue from $2.62B in 2022 to $5.22B in 2025, while net income swung from a $44M loss to $1.53B. That is not accounting vapor. The latest four quarters sum to roughly $5.37B of revenue and $1.74B of net income, implying a trailing P/E closer to 14x on today’s $24.24B market cap, and that multiple sits on a balance sheet with $2.80B of cash against just $499M of debt. A business producing 29%-plus annual net margins, 21% ROIC, and $1.19B of free cash flow after nearly $870M of capex should not be treated like a speculative concept stock. What stands out most is that profitability has remained strong even as revenue scaled: gross margin was 40.6% in 2025, operating margin 30.6%, and recent quarterly net margins have run 28.6%, 31.0%, 33.2%, and 40.0%. That is a manufacturer with unusual structural economics, not a commodity player earning a fleeting spread.
The quarter pattern is also better than the “decelerating” label suggests. Yes, revenue stepped down from $1.68B in 4Q25 to $1.04B and $1.06B in the first two quarters of 2026, but sequential lumpiness is normal in utility-scale solar. More important is the year-over-year base: 1Q26 revenue of $1.04B was up about 23% from 1Q25’s $845M, and 2Q26 revenue of $1.06B was only down about 4% from 2Q25’s $1.10B while net income jumped 24% to $423M and margin expanded to 40%. That tells me mix, pricing, manufacturing credits, or execution are offsetting volume unevenness. The market appears to be paying roughly 4.6x sales and 10.3x EV/EBITDA for a capital-intensive domestic manufacturer with net cash, visible earnings, and superior margins. Those are not giveaway multiples, but they are far from euphoric if one believes even a mid-teens earnings base is sustainable. Book value is $9.54B, so 2.5x book for a business earning 16% ROE and still building capacity is reasonable.
Where I part company with the bullish reflex is on durability, not on current quality. The data scream that 2023-2026 has been a step-change period aided by exceptional industry structure and policy support. The jump from 2.7% gross margin in 2022 to 39%-44% gross margins thereafter is too violent to assume as a normalized through-cycle state. A smart investor should assume today’s earnings are elevated by favorable contract timing, domestic content premiums, and subsidy architecture, and that some of the 30%-40% net margins will compress as supply catches up, projects slip, or policy economics change. If I haircut annual earnings from the current ~$1.7B run rate down toward $1.3B-$1.4B, the stock is trading closer to 17x-19x normalized earnings, which is no bargain for a cyclical manufacturer. So the right read is not “massively overvalued” and not “screaming cheap”; it is that the market is assigning a moderate premium to unusually high-quality current earnings while discounting some fade.
The strongest case against my view is straightforward: quarterly revenue has likely peaked for now, recent growth has slowed from the 2023-2025 surge, the company operates in a politically constructed moat, and valuation on sales/book is not low for an industrial. If someone argues that 2025’s $1.53B net income was peak and 2026 margins around 33%-40% are unsustainably rich, they can reasonably say the stock deserves a low-teens multiple on trough-ish or normalized earnings, which would land closer to $170-$190 than $225. They would also point out that free cash flow quality was flagged poor and that capex remains heavy, so one should not capitalize peak earnings too aggressively. I weigh those points differently because the cash generation is still solid even after expansion spend, leverage is minimal, and the consistency of profitability across the last six quarters makes it hard to dismiss as a one-quarter subsidy mirage.
What would change my mind is specific evidence that earnings power is rolling over rather than merely smoothing through project timing. If the next two quarters show revenue stuck near $1.0B but net margin falling back into the mid-teens or below, the thesis of durable structural profitability is broken and I would become bearish quickly. Likewise, if annual free cash flow drops materially below $800M without a commensurate capacity payoff, or if policy changes threaten the economics underpinning recent margin expansion, the appropriate multiple shrinks. On the other hand, if First Solar can hold revenue above roughly $5B annualized while sustaining operating margins above 25% and free cash flow above $1B, then today’s valuation is not demanding and a price north of $250 would be supportable.
Grok Reading
The numbers tell a straightforward industrial compounding story that the valuation engine is badly misreading. First Solar went from $3.32B revenue and $831M net income in 2023 to $5.22B and $1.53B in 2025, a 57% top-line lift and nearly doubled earnings in two years, with revenue CAGR still running 25% and earnings CAGR 36%. Net margin expanded to 29.3%, operating margin to 30.6%, and gross margin to 40.6%—levels that look nothing like a commoditized panel maker scraping for share. The balance sheet is pristine: $2.80B cash against $499M debt, debt-to-equity of 0.05, current ratio 2.67. Operating cash flow of $2.06B funded $870M of capex and still left $1.19B of free cash flow. At $225.56 the stock trades 15.9x trailing earnings, 10.3x EV/EBITDA, and 2.5x book. Those are not bubble multiples for a 25%-growth manufacturer printing 30% net margins and mid-teens ROE with a net cash fortress. The $37 DCF “fair value” implying an 83% overvaluation and a ~2.6x earnings multiple is simply broken; it is incompatible with the income statement and cash flow the model itself was fed.
What stands out quarterly is both the margin trajectory and the cadence. Net margin hit 40% in the June 2026 quarter on $1.06B revenue and 33% the quarter before—still elite—while the second half of 2025 printed the volume peaks ($1.59B and $1.68B). First-half 2026 revenue is running below those H2 peaks, consistent with the flagged decelerating quarterly trend and typical solar seasonality, not a collapse. Annual growth remains firmly positive. ROIC at 21% says the incremental capital being deployed into U.S. manufacturing capacity is earning well above cost of capital. Insider activity is pure award noise; it neither confirms nor contradicts. The real story in the raw data is a high-return, policy-supported thin-film manufacturer that has already crossed into mature-earner economics while still growing mid-twenties, priced as if those economics are about to vanish.
The strongest case against this read is the one the narrative and valuation layers lean on hard. Solar manufacturing is policy-exposed; a large slice of the margin structure and domestic buildout economics rests on IRA manufacturing credits and related U.S. content preferences. If those are repealed, scaled back, or allowed to sunset without replacement, both volume and unit economics compress. Crystalline-silicon competitors remain cheaper on a pure module basis, and any normalization of supply chains or Chinese export dynamics can pressure ASP and share. The secondary flag on poor cash-flow quality, even with $1.19B of reported FCF, deserves scrutiny—working-capital swings and credit timing can make earnings look cleaner than cash conversion over shorter windows. Revenue confidence is only “adequate” with a decelerating quarterly trend, and the macro environment is labeled headwinds. A smart opponent also notes that 16x earnings for a cyclical industrial with political residual risk is not automatically cheap once you strip the energy-transition story; if growth settles to high-single-digits and margins mean-revert toward the mid-teens, the multiple compresses and the stock can sit or fall for years even if the business remains profitable. I weigh those risks as real but already partially in the multiple: the market is not paying 30x or 40x for the narrative, it is paying roughly 16x for demonstrated 30% margins and 25% growth. That is skepticism priced in, not euphoria.
I would flip if trailing net margins sustainably fell below 18–20% without an offsetting volume surge, if full-year revenue growth dropped into the single digits for two consecutive years, or if concrete legislative action removed the bulk of domestic manufacturing credit support. A clean print of another $5.5B-plus revenue year with FCF holding above $1B and margins still above 25% would, conversely, make the undervaluation case more aggressive.
Big-3 Panel — where each AI stands
Advanced Analysis Forensic deep-dive · separate lenses
The business is in visibly strong shape: revenue has climbed from $2.92B (2021) to $5.22B (2025), gross margin expanded from 25% to 40.6%, and operating margin from 20.1% to 30.6%. Net income scaled from $468M to $1.53B over the same window. The 2022 collapse (GM 2.7%, -$44M NI) shows the business is cyclical and sensitive to input/logistics shocks, but the recovery has been decisive and margins now sit at industry-leading levels for a hardware manufacturer.
Verify before trusting this (5)
- Customer concentration and contracted backlog duration in the 10-K
- Exposure to IRA 45X manufacturing credits — how much of 2024-2025 profit is credit-driven
- Capex trajectory and whether 2025 FCF inflection is sustainable or a one-year working-capital effect
- Tariff/anti-dumping regime dependence and any policy sunset risk
- Warranty reserves and product-quality accruals given series 7 module ramp
The composite fair value of $36.90 is almost certainly a runaway artifact - the DCF at $3.45 is nonsensical for a self-funding manufacturer with a fortress balance sheet, and even the EPV floor of $103.80 sits less than half of today's $225.56. Even if you generously double the EPV floor to credit IRA-era earnings power and a strong quality grade, you land near $200-220 - roughly today's price, not below it. There is no visible margin of safety.
Verify before trusting this (4)
- Durability of 45X credit realization in 2025-26 cash flows and any legislative risk to IRA
- Backlog ASPs and cancellation/repricing terms as module oversupply persists
- FCF conversion trend - is 2024's catch-up repeatable or a working-capital swing
- Booking pace beyond 2027 to test the 'sold out through decade' claim
The macro tape is mildly risk-on with VIX at 14.3 and indices near highs, and with a 1.75 beta FSLR gets an amplified lift from that backdrop rather than the muted push a defensive name would feel. The active narrative is a mission-driven, high-cult energy-transition story running with strong intensity, and the last 72 hours refreshed it: solar just passed wind in US generation, headlines note solar is half of all new grid capacity, and, most concretely, UBS and BNP framed the new US solar tariffs (with a $0.38/W minimum import price) as a direct competitive gift to First Solar. That is exactly the kind of narrative-confirming catalyst that keeps a cult story bid. Against that, the bear framing that the stock trades far above DCF on hype is a real latent headwind, and one piece flagged FSLR-adjacent names on a downbeat-forecast list. But sentiment is about pressure, not merit: right now the story is being validated by policy and generation data, and the tape is not fighting it. Net, the non-fundamental push on THIS name is clearly upward, though its durability depends on policy continuity.
Verify before trusting this (5)
- Whether the tariff/minimum-import-price rule is finalized as previewed or watered down
- Any shift in IRA or federal solar policy tone that could crack the permanence assumption
- Analyst target revisions post-tariff news - do buy-side price targets follow UBS/BNP higher
- Sector ETF (TAN) flows and breadth confirming the narrative is broadening, not just FSLR-specific
- VIX break above 18-20 or a risk-off turn that would punish beta 1.75 names disproportionately
First Solar sells a physical commodity whose cost is dominated by glass, semiconductor deposition, depreciation and energy — none of it information-processing, so AI cannot substitute the product, disintermediate the sale, or let customers self-serve; utilities will not build modules. AI therefore enters through three narrow channels: a genuine demand shock as datacenter load makes deployable generation scarce (favorable, and the dominant channel), modest COGS relief from yield/uptime optimization in a capex-heavy cost stack (small, and likely competed away in a price-taking module market), and an R&D leveling effect where cheap simulation and autonomous experimentation shorten the empirical advantage of a proprietary thin-film platform against a far larger silicon and perovskite research pool (unfavorable, slow-burning). Net: AI is an amplifier of the volume thesis and a quiet eroder of the technology-differentiation thesis; it does nothing about the actual bear case, which is Chinese silicon oversupply and policy dependence.
None surfaced.
Verify before trusting this (8)
- Data-center-linked PPA volumes
- US load growth forecasts
- Interconnection queue durations
- Storage pairing attach rates
- CdTe vs silicon efficiency gap
- Perovskite tandem pilot lines
- CuRe/bifacial roadmap milestones
- Warranty claim trends
The world is adding electricity demand faster than in two decades (electrification, data-center load), which underwrites solar volume broadly. But FSLR's specific exposure is narrower than the theme: it is a US utility-scale module supplier whose economics rest on manufacturing credits, trade duties and tax-equity-financed customers. That makes legislation and rates, not sunlight, the swing variables. Near term the policy step-down is a tailwind — developers pull projects forward to safe-harbor before credits lapse — which flatters 2026-27 shipments and creates an air pocket after. Globally, structural polysilicon oversupply keeps module pricing deflationary everywhere tariffs don't reach, so FSLR's addressable growth is domestic by construction. Macro backdrop is a mild headwind: a 4.63% 10y raises project hurdle rates just as PPA pricing faces buyer resistance.
Prediction unavailable. valuation-synthesis has no result for FSLR — the prediction needs its fair-value anchors.