Multibagger Selection: GPT-5.6 Sol's picks
GPT-5.6 Sol picked LUNR, RCAT, POET, ONDS, AEVA with a twelve-month thesis, measured against SPY at each checkpoint.
| Experiment | Multibagger Selection |
|---|---|
| Model | GPT-5.6 Sol |
| 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 | LUNR | — | 84% | — | — |
| 2 | RCAT | — | 81% | — | — |
| 3 | POET | — | 75% | — | — |
| 4 | ONDS | — | 79% | — | — |
| 5 | AEVA | — | 71% | — | — |
Checkpoints
| Checkpoint | Portfolio return | Return vs SPY | Best peak multiple | Picks reaching 2x |
|---|---|---|---|---|
| 1 week | -3.24% | -2.58% | 1.07x | 0 |
| 1 month | -27.86% | -27.12% | 1.07x | 0 |
The model's market view
The strongest sub-$5 billion multibagger candidates are concentrated in businesses where a small current revenue base is crossing into production scale: defense drones, space infrastructure, autonomous systems and AI optical interconnects. I rank operating evidence and balance-sheet survivability above thematic excitement. The experiment is explicitly measuring 12-month multibagger discovery rather than short-term performance, with checkpoints through September 24, 2027. :contentReference[oaicite:0]{index=0}
Why each pick
LUNR
Intuitive Machines has the strongest combination of real revenue scale, backlog visibility and identifiable 12-month catalysts. Q2 2026 revenue reached $206.2 million, up 310% year over year, TTM revenue reached $490.1 million, and quarter-end backlog expanded to approximately $1.8 billion after $920 million of Q2 awards plus additional Q3 awards. National-security revenue increased from 3% to 30% of Q2 revenue, materially broadening the business beyond lunar missions. The principal reason a 3x scenario is economically conceivable is that the business is already producing hundreds of millions of dollars of revenue rather than depending on a pre-revenue technology transition. Sources: https://investors.intuitivemachines.com/news-releases/news-release-details/intuitive-machines-reports-second-quarter-2026-financial-results and https://stockanalysis.com/stocks/lunr/
RCAT
Red Cat is undergoing an unusually sharp transition from development-stage defense-drone supplier to volume producer. Q2 2026 revenue increased 527% year over year to $20.2 million, six-month revenue increased 636% to $35.7 million, and Q2 gross margin improved to 16.1% from 11.6% a year earlier and sequentially from Q1. Revenue growth is being driven by U.S. Army drone deliveries and initial Japan Ground Self-Defense Force deliveries rather than an unverified future market. The approximately $1.01 billion market capitalization remains small enough that successful production scaling can materially change the valuation. Sources: https://ir.redcatholdings.com/sec-filings/all-sec-filings/content/0001628280-26-054368/rcat-20260630.htm and https://ir.redcatholdings.com/sec-filings/all-sec-filings/content/0001628280-26-054364/rcat-20260806xexx991a.htm
POET
POET is the highest technological and commercialization asymmetry in the group but also one of the most speculative. TTM revenue is only $1.71 million, yet the company has moved from sampling toward production orders for AI-data-center optical engines. It has an initial $50 million Lumilens purchase order inside a framework targeting more than $500 million over five years, plus an earlier $5+ million production order targeted for shipment in the second half of 2026. The balance sheet is unusually strong for this stage, with nearly $800 million of cash and short-term investments against less than $8 million of debt. Sources: https://www.poet-technologies.com/blog/faq-production-orders-cash-position-and-the-path-to-greater-revenue and https://stockanalysis.com/stocks/poet/statistics/
ONDS
Ondas has the strongest near-term financial acceleration in the set but starts from the largest market capitalization, which reduces multibagger asymmetry. Q2 2026 revenue was $83.8 million, 67% above Q1 and roughly 13 times Q2 2025. The company reported approximately $613 million of backlog, or $757 million pro forma including acquisitions, and raised its 2026 revenue target to $525-$550 million. It also held roughly $1.4 billion of cash and short-term investments before completing recent acquisitions. Sources: https://ir.ondas.com/press-releases/detail/326/ondas-posts-record-q2-2026-revenue-of-83-8-million-on and https://ir.ondas.com/all-sec-filings/content/0001193125-26-347973/onds-ex99_2.htm
AEVA
Aeva combines a small enough market capitalization for genuine multibagger asymmetry with several externally validated commercialization programs. NVIDIA selected Aeva's 4D LiDAR as the reference sensor for DRIVE Hyperion, a top European passenger OEM awarded Aeva an exclusive global production contract outside China, Daimler Truck is progressing toward scaled production, and Aeva opened a new AI-data-center optical-connectivity business with an initial hyperscaler deployment targeted for the second half of 2027. Current revenue remains small, so the thesis is explicitly about contract conversion rather than present earnings. Sources: https://investors.aeva.com/news-releases/news-release-details/aeva-and-nvidia-integrate-4d-lidar-reference-sensor-within and https://investors.aeva.com/news-releases/news-release-details/aeva-launches-optical-connectivity-business-signs-first-customer
Other models on the same run
| Model | Picks |
|---|---|
| Claude Opus 5 | ICHR, OUST, RUM, USAR, EMAT |
| Muse Spark 1.1 | KPLT, INDP, BNC, CHPT, SWVL |
| Gemini 1.5 Pro | RCAT, LUNR, RDW, SMR, ENVX |
| Grok 4.5 | VRDN, REPL, NRIX, AEVA, SERV |
| DeepSeek | INOD, UMAC, CAPR, NUAI, PDYN |
ForecastNest provides AI-generated market analysis and quantitative research for informational and educational purposes only. Nothing here constitutes financial, investment, legal, or tax advice.