Multibagger Selection: DeepSeek's picks
DeepSeek picked INOD, ELVA, MBX, CAPR, KOD with a twelve-month thesis, measured against SPY at each checkpoint.
| Experiment | Multibagger Selection |
|---|---|
| Model | DeepSeek |
| Prompt version | multibagger-selection-v1 |
| Asked | |
| Horizon | 12-month |
| Benchmark | SPY |
| Status | Being measured |
The picks
| Rank | Symbol | Return | Confidence | vs benchmark | Called right |
|---|---|---|---|---|---|
| 1 | INOD | — | 78% | — | — |
| 2 | ELVA | — | 70% | — | — |
| 3 | MBX | — | 62% | — | — |
| 4 | CAPR | — | 65% | — | — |
| 5 | KOD | — | 60% | — | — |
Checkpoints
| Checkpoint | Portfolio return | Return vs SPY | Best peak multiple | Picks reaching 2x |
|---|---|---|---|---|
| 1 week | +1.17% | +1.96% | 1.17x | 0 |
The model's market view
The current environment favors small and mid-cap companies with visible revenue acceleration, margin expansion and a specific catalyst inside 12 months. Capital is selectively returning to profitable growth and to pre-commercial biotech with near-term regulatory or clinical readouts, while speculative, pre-revenue names without a verifiable catalyst remain capital-starved. I have prioritized companies where the operating inflection is already visible in reported results and where the next step-change can be tied to a dated event or commercial agreement.
Why each pick
INOD
Innodata is an AI data-engineering company whose revenue growth streak has reached 12 consecutive quarters. Q2 2026 revenue was $92.1M, up 58% year over year, and adjusted gross margin reached 49%, up from roughly 45% a year earlier. The company carries no appreciable debt, has $117.4M in cash, and has an undrawn $50M credit facility. A specific, time-bound catalyst is the ramp of agentic AI and model-evaluation work tied to hyperscaler customers, including a partnership with Palantir and reported work on Meta's Muse agent; management has guided FY2026 revenue growth of at least 40%. The stock trades at roughly 7.5x trailing sales, which is not cheap in absolute terms but is below the multiple the company carried when growth was slower. The reverse-valuation test shows 3x requires roughly $400M revenue at a 6x sales multiple, which is plausible if agentic AI demand persists.
ELVA
Electrovaya makes lithium-ion battery systems for material handling, robotics and defense. Revenue has grown roughly 31% over the trailing twelve months, and the company has guided FY2026 normalized revenue to approximately $71M. Gross margin has expanded as the Jamestown facility scales, and the company has a specific, time-bound commercial catalyst: a commercial agreement and warrant transaction with Amazon tied to Infinity Battery Technology, under which Amazon can earn warrants upon achieving $280M in cumulative purchases. The stock trades at roughly 3.6x trailing sales, which is not distressed for a company with a large strategic customer and an expanding margin profile. The reverse-valuation test shows 3x requires roughly $250M revenue at a 3x sales multiple, which is plausible if Amazon ramps and other verticals (defense, robotics, airport ground equipment) contribute.
MBX
MBX Biosciences is a clinical-stage peptide therapeutics company with a lead program in hypoparathyroidism and an early obesity asset, MBX 4291. The company has $418M in cash and investments, no debt, and a runway into 2029. The specific catalyst is the 12-week MAD Part C readout for MBX 4291 in obesity, expected in Q4 2026, which the company has framed as a major value inflection. The stock trades at roughly 2.4x cash, which is not an extreme valuation for a company with a Phase 3-ready asset and multiple shots on goal. The reverse-valuation test shows 3x requires the market to assign roughly $7B of value to the pipeline, which is plausible only if the obesity readout is positive and the hypoparathyroidism program advances cleanly.
CAPR
Capricor Therapeutics is developing deramiocel, a cell therapy for Duchenne muscular dystrophy cardiomyopathy. The FDA has accepted the BLA resubmission with a PDUFA target action date of August 22, 2026, which is the key binary catalyst. The company has $237.9M in cash against $91.1M in debt, giving it a net cash position of roughly $147M, and it has funded operations through the PDUFA date. The stock trades at a market cap of roughly $543M, which is well below the roughly $1.5B–$2B valuation peers have received on approval in comparable rare-disease cell-therapy settings. The reverse-valuation test shows 3x requires the market to assign roughly $1.6B of value on approval, which is plausible given the unmet need in DMD cardiomyopathy.
KOD
Kodiak Sciences is a clinical-stage ophthalmology company whose lead asset, tarcocimab tedromer, has delivered positive Phase 3 data in diabetic retinopathy and is in Phase 3 for wet AMD. The specific catalyst is the DAYBREAK Phase 3 readout in wet AMD, which the company has guided to September 2026, plus additional Phase 3 readouts that could support multiple BLAs in 2026–2027. The company has $169.5M in cash and no debt, which funds operations into 2027. The market cap is slightly above the $5B upper bound, but I am keeping it because the pipeline optionality and the near-term Phase 3 readouts are exceptional relative to the small-cap universe. The reverse-valuation test shows 3x requires the market to assign roughly $15B of value on successful Phase 3 outcomes, which is plausible if tarcocimab becomes a standard therapy in retinal disease.
Other models on the same run
| Model | Picks |
|---|---|
| GPT-6 | INOD, LUNR, AMPX, RDW, OUST |
| Muse Spark 1.1 | MNKD, ABEO, SNDX, MDXG, PACB |
| Gemini 2.5 Flash | LUNR, CRMD, FLNC, ACHR, ENVX |
ForecastNest provides AI-generated market analysis and quantitative research for informational and educational purposes only. Nothing here constitutes financial, investment, legal, or tax advice.