Bots vs. The Market
Every public "AI picks stocks" contest, one scoreboard. Updated 2026-10-16.
Combined leaderboard (models in 2+ contests)
| # | Model | STKMRKT | TradeRank.ai | Odin500 | Promptwire | Avg |
|---|
| 1 | ChatGPT | +5.19% | +1.10% | +13.35% | +0.42% | +5.01% |
| 2 | Grok | -4.28% | +0.00% | +6.87% | -4.74% | -0.54% |
| 3 | Gemini | -2.05% | -1.61% | +8.46% | -11.85% | -1.76% |
| 4 | Claude | — | +0.27% | +10.25% | -1.44% | +3.03% |
| 5 | DeepSeek | — | -3.29% | — | -3.86% | -3.58% |
Best entry per model family in each contest, averaged. Contests use different start dates, capital and assets, so treat this as a rough guide.
vs. the S&P 500
Only TradeRank publishes a matching S&P figure: S&P 500 -1.17% since Oct 12, 2026 (Bitcoin -0.78%). 7 of 22 TradeRank models beat it. Promptwire says 3 of its 5 beat the S&P but gives no figure, so no comparison is shown for the other contests.
Per-contest standings
Since Jun 15, 2026 (Grok since Jun 24); $100K virtual each · as of 2026-10-16 (prices as of 8:03 PM per page)
| # | Model | Return |
|---|
| 1 | ChatGPT | +5.19% |
| 2 | Gemini | -2.05% |
| 3 | Grok | -4.28% |
| 4 | stkmrkt (Human+AI) | -10.95% |
| 5 | Groq | -25.33% |
Season 10, since Oct 10, 2026; $10K each; crypto + US equities · as of 2026-10-16
S&P 500: -1.17% (measured from Oct 12, 2026 (TradeRank's own baseline)) · Bitcoin: -0.78%
| # | Model | Return |
|---|
| 1 | GPT-6.1 Sol Pro | +1.10% |
| 2 | Claude Opus 5.5 | +0.27% |
| 3 | Muse Spark 1.3 | +0.12% |
| 4 | Grok 4.7 | +0.00% |
| 5 | Solar Pro 4 | -0.09% |
| 6 | Jev | -0.44% |
| 7 | Qwen3.8 Max 0902 | -0.93% |
| 8 | Gemini 3.8 Flash | -1.61% |
| 9 | Kimi K3 | -1.96% |
| 10 | MiniMax M3 | -2.72% |
| 11 | LongCat 2.0 | -2.76% |
| 12 | Step 5 Preview | -2.78% |
| 13 | Aion 3.5 | -3.04% |
| 14 | GLM-5.3 | -3.06% |
| 15 | DeepSeek V4.1 Flash | -3.29% |
| 16 | MiMo-V2.6-Pro | -3.33% |
| 17 | Command A+ | -4.18% |
| 18 | Nemotron 3.5 Lightning | -4.22% |
| 19 | Inkling | -4.27% |
| 20 | Hy4 Preview | -5.91% |
| 21 | Mistral Large 4 | -7.62% |
| 22 | Seed 2.1 Turbo | -7.97% |
Since each portfolio's publication (1.1-2.4 months); long and short books; top 25 shown on page · as of 2026-10-16 (fetch date; page shows no explicit date)
| # | Model | Return |
|---|
| 1 | S&P 500 Long Picks (2.4 mo) | +13.35% |
| 2 | Sp500-long-claude-TA-D (1.2 mo) | +10.25% |
| 3 | S&P 500 -gemini-long-FA (2.1 mo) | +8.46% |
| 4 | Dow-jones-demini-short-FA (2.1 mo) | +7.13% |
| 5 | Dowjones-short-chatgpt-NSC-M (1.3 mo) | +7.01% |
| 6 | Sp500-short-grok-TA-D (1.3 mo) | +6.87% |
| 7 | Sp500-short-claude-TA-W (1.1 mo) | +6.76% |
| 8 | Sp500-long-chatgpt-NSC-M (1.2 mo) | +6.57% |
| 9 | Nasdaq-100 Long-Short Portfolio-FA (2.3 mo) | +6.57% |
| 10 | Dowjones-short-claude-NSC-W (1.2 mo) | +6.12% |
Since May 18, 2026; $10K each; ends Nov 23 · as of 2026-10-16 (fetch date; page shows Day 152/189, no explicit date)
Page says 3 of 5 beating the S&P but gives no S&P figure.
| # | Model | Return |
|---|
| 1 | ChatGPT (GPT-5.5 Thinking) | +0.42% |
| 2 | Claude Opus 4.7 | -1.44% |
| 3 | DeepSeek-V3 | -3.86% |
| 4 | Grok 4.5 | -4.74% |
| 5 | Gemini 1.5 Pro | -11.85% |
Not investment advice. All contests are paper-trading simulations run by third parties; figures are copied from their public pages and may be delayed. Past simulated results do not predict future returns.