Forensic Analysis · Healthcare / Pharmaceuticals / Biotech · as of Sep 25, 2026
Alumis Inc. (ALMS)
A forensic read on Alumis Inc. built from its complete SEC filings — financial-health screens, earnings quality, red flags and a price-aware rating. Reproducible math, not opinion.
F · Poor — capital at risk
Forensic grade
Distress
Financial health
-1.5
Distress distance
Clean
Earnings quality
5
Forensic signals
-2.0
P / E (ttm)
-80.8%
ROE
$963M
Market cap
0.00%
Dividend yield
-37.7%
Revenue growth
This free snapshot doesn't read filing or proxy text, so the auditor-quality and governance checks that can lower a grade were not evaluated (red flags, auditor quality, say on pay, board independence, going concern, valuation). Those checks only ever lower a grade, so the full analysis can land below the grade above but never above it — treat it as the most favourable reading, not the settled one.
Alumis Inc. earns an F (Poor — capital at risk) forensic quality grade, and its balance-sheet distress test reads -1.5, placing it in the Distress zone. 5 forensic signals were flagged in its latest SEC filings, led by accruals ratio (% of noa).
What the filings flag
+80.1%
FY2024→FY2025
Accruals ratio (% of NOA).Net operating assets grew +80.1% relative to their own average in FY2025 — scaled by NOA itself, not by total assets, which is what the Forensic Screens card's own accrual row divides by. Accruals are building sharply — a large slice of profit sits in operating assets, not cash. An accrual is a claim that still has to be collected or written down, so the gap resolves in a later period whichever way it goes. A cash-flow measure on the same base agrees: reported earnings ran ahead of operating cash by 84% of net operating assets, against an accruals ratio of 80.1%. The two are computed differently: the accruals ratio is the change in net operating assets over average net operating assets, while the cash-flow figure is net income less operating cash flow over that same average.
FCF ($370M)
FY2025
Shareholder returns.Returned $17,000 to shareholders (buybacks + dividends) in FY2025, but free cash flow was ($370M) after capex — there was no free cash flow to fund the payout from at all, and operating cash flow itself was negative or zero that year too. The entire return is coming from debt or cash reserves, not cash the business itself generated — a harder case than returns merely running ahead of free cash flow, since here there was none to run ahead of.
-$624.6M
FY2024–FY2025
Cash burn vs. reported loss.Over FY2024–FY2025, the company reported a cumulative net loss of $537.6M against operating cash flow of -$624.6M. Cash burn ran heavier than the reported loss — something outside net income (working capital, a cash item not in the P&L) is consuming cash faster than the loss alone implies.
n/m (stock split)
FY2024–FY2025
Share count (stock split).
Key fundamentals
Latest Revenue$24.1M
Net Margin-1011.7%
Free Cash Flow-$370.2M
Return on Equity-80.8%
The forensic grade and screens above are free — no account needed. Want the AI investment read on top — the 0–100 rating, thesis, bull-vs-bear, red flags and the 12-month scenario, written from Alumis Inc.'s actual 10-K/10-Q/8-K filings?
Data from SEC EDGAR public filings · metrics as of Sep 25, 2026. Forensic signals flag probability, not certainty.
Alumis Inc. (ALMS) Stock — Forensic Analysis, Red Flags & Rating | Stockonomy · Stockonomy
Diluted share count changed +200% over the last 1 year to FY2025, but that includes a large one-time change around FY2025 consistent with a stock split or reverse split, not gradual buybacks or issuance — a split changes the count with NO effect on any holder's proportional ownership, so the raw +200.0%/yr figure isn't a real buyback/dilution read here.
181% of rev
FY2025
Stock-based comp load.Stock-based compensation ran 181% of revenue in FY2025 — about $0.51 per diluted share. It is a real cost, but it is not a cash cost — no cash left the business, which is why operating cash flow adds it back. Where a compensation charge lands instead is the share count, and this filer's count is not on file in enough years to say how the count moved.