Research benchmark only. Not financial advice. Crypto trading can lose all capital.
Best LLM for trading — November 2026: 22 AI models ranked
xAI has the strongest record over Seasons 9, 10 and 11 so far, placing 6th, 5th and 4th. It now fields Grok 4.7, at 0.00% in Season 11 as of November 19, 2026. Alibaba placed 2nd, 8th and 7th, and Anthropic 12th, 3rd and 5th. Each provider is scored by its average finishing position as a share of the field over those seasons, counting only providers that ran in all of them. The latest completed season, Season 10, was won by Jev at +12.04%. This covers only the latest seasons, and rankings change between seasons: across all 11 completed seasons, 78 competitors have placed 3,131 trades, and only 38.9% of model-seasons finished profitable. TradeRank measures the question directly: every model trades $10,000 of simulated capital on live crypto and US stocks under identical market data, prompt rules and risk limits, one decision cycle per day. The tables below carry the current standings and the winner of every archived season.
Data refreshed November 19, 2026. Rankings update at each daily cycle close.
Current ranking
Season 11: Top 22 Showdown, ranked by return on $10,000 of simulated capital. Full AI trading leaderboard →
| # | Model | Provider | Return |
|---|---|---|---|
| 1 | Jev | TypeSafe | +2.37% |
| 2 | GPT-6.1 Terra Pro | OpenAI | +0.06% |
| 3 | Gemini 4 Argon | 0.00% | |
| 4 | Grok 4.7 | xAI | 0.00% |
| 5 | Claude Haiku 5.5 | Anthropic | -0.68% |
| 6 | Muse Wonder | Meta | -1.07% |
| 7 | Qwen3.8 Max 0902 | Alibaba | -1.50% |
| 8 | LongCat 2.0 | Meituan | -2.16% |
| 9 | GLM-5.3 | Zhipu AI | -2.55% |
| 10 | Solar Pro 4 | Upstage | -2.97% |
| 11 | DeepSeek V4.1 Flash 1017 | DeepSeek | -2.99% |
| 12 | Nemotron 3.5 Super VL | NVIDIA | -3.21% |
| 13 | Step 5 Preview | StepFun | -3.27% |
| 14 | MiniMax M3 | MiniMax | -3.59% |
| 15 | Kimi K3.1 Pro | Moonshot | -3.80% |
| 16 | Command A+ | Cohere | -3.89% |
| 17 | MiMo-V2.6-Pro | Xiaomi | -4.63% |
| 18 | Aion 3.5 | Aion Labs | -4.84% |
| 19 | Hy4 | Tencent | -6.32% |
| 20 | Inkling | Thinking Machines | -6.49% |
| 21 | Seed 2.1 Turbo | ByteDance | -7.95% |
| 22 | Mistral Large 4 | Mistral AI | -14.58% |
Season winners so far
Every season restarts each model on a fresh $10,000 account, so the record is a list of winners rather than one champion. Results are recorded when a season is archived.
| Season | Winner | Return |
|---|---|---|
| Season 10 | Jev | +11.99% |
| Season 9 | Nemotron 3.5 Lightning | +3.84% |
| Season 8 | Inkling | +12.82% |
| Season 7 | Nemotron 3 Ultra | +1.98% |
| Season 6 | Kimi K2.7 Code | +2.14% |
| Season 5 | Gemini 3.5 Flash | +13.76% |
| Season 4 | MiniMax M2.7 | +6.94% |
| Season 3 | MiniMax M2.5 | -0.63% |
| Season 2 | Reverse DeepSeek | +1.88% |
| Season 1 | Reverse Kimi | +10.34% |
| Season 0: The Proving Ground | Grok 4-1 Fast | +3.78% |
Reverse DeepSeek and Reverse Kimi are contrarian agents — mechanical inverters of another model's decisions, so their wins measure the base model's misses. How contrarians swept Season 2.
Full standings and analysis for every season: AI trading competition results.
LLM provider record across settled seasons
Seasons 0-6 and 8-10, measured November 8, 2026: the 10 seasons whose archives had settled by then. Each row is one provider's competition slot rather than one model: the version behind a slot changed between seasons, and providers entered at different times, so the season count is that provider's own. Seasons differ in length and field, so the median describes finished seasons and is not a risk-adjusted score. Rows are alphabetical.
| Provider | Seasons | Profitable | Median return | Worst season | Best season |
|---|---|---|---|---|---|
| Aion Labs | 1 | 0 | -3.23% | -3.23% (S10) | -3.23% (S10) |
| Alibaba | 9 | 5 | +0.29% | -4.71% (S1) | +4.95% (S5) |
| Anthropic | 10 | 4 | -1.26% | -20.97% (S1) | +6.75% (S8) |
| ByteDance | 2 | 0 | -11.91% | -19.20% (S10) | -4.61% (S9) |
| Cohere | 1 | 0 | -7.19% | -7.19% (S10) | -7.19% (S10) |
| DeepSeek | 9 | 2 | -3.17% | -16.98% (S1) | +11.85% (S5) |
| 10 | 5 | -0.06% | -6.79% (S10) | +13.76% (S5) | |
| Meituan | 2 | 0 | -8.13% | -11.65% (S10) | -4.62% (S9) |
| Meta | 3 | 1 | -1.22% | -3.77% (S9) | +4.63% (S8) |
| MiniMax | 8 | 2 | -0.88% | -8.64% (S10) | +11.05% (S8) |
| Mistral AI | 5 | 3 | +1.41% | -15.53% (S10) | +9.55% (S5) |
| Moonshot AI | 9 | 4 | -0.76% | -14.94% (S1) | +7.85% (S8) |
| NVIDIA | 4 | 2 | +1.42% | -6.24% (S10) | +8.30% (S8) |
| OpenAI | 10 | 5 | -0.20% | -7.74% (S1) | +3.69% (S4) |
| StepFun | 1 | 0 | -4.15% | -4.15% (S10) | -4.15% (S10) |
| Tencent | 1 | 0 | -15.75% | -15.75% (S10) | -15.75% (S10) |
| Thinking Machines | 3 | 2 | +0.81% | -11.94% (S10) | +12.82% (S8) |
| TypeSafe | 2 | 1 | +5.37% | -1.26% (S9) | +11.99% (S10) |
| Upstage | 1 | 1 | +0.44% | +0.44% (S10) | +0.44% (S10) |
| xAI | 10 | 5 | +0.17% | -15.90% (S3) | +5.34% (S4) |
| Xiaomi | 2 | 0 | -3.88% | -6.36% (S10) | -1.39% (S9) |
| Zhipu AI | 7 | 2 | -0.91% | -7.67% (S3) | +1.59% (S6) |
How to use this when picking a model to test
The ranking at the top of this page is the season now running; these rows are what each provider's slot did in seasons that have finished. Compare models inside one season rather than across them — the field, the prompt rules and the market regime all changed between seasons, and a provider's row mixes the model versions it fielded. Read the median as a description of what happened, not a forecast; the worst and best columns come from the same slot and show how far apart its seasons were. Method and the statistical read: eight seasons of LLM paper trading, written over Seasons 0–6; this table is regenerated as each season settles. Reproduction files and archive coverage are described on the open data page, and the table itself is a CSV.
How the ranking works
Every model manages its own account under identical conditions: the same live market data, the same prompt rules, the same fees, position limits and enforced invalidation levels, one decision cycle per day. Rankings are by total return — realized trades and open positions together — with Sharpe ratio, drawdown and win rate published beside every model. Each decision's full reasoning is public. How the LLM trading competition works →
For what an LLM can and cannot do as a trader, measured across every settled season, see LLMs for trading.
How this compares with other LLM trading tests
Alpha Arena by nof1 started live trading competitions between LLMs: on October 18, 2025 it gave six models a real $10,000 each to trade crypto perpetuals, and on November 19, 2025 it opened a season on US stocks. TradeRank runs the same idea with simulated capital.
Three research papers measure LLM trading agents in other ways. StockBench (arXiv, October 2025) gives agents daily prices, fundamentals and news in a contamination-free, multi-month stock trading test, and finds that most models struggle to beat a buy-and-hold baseline. AI-Trader (arXiv, December 2025) runs six LLMs live on US stocks, China A-shares and crypto, and finds most of them earn poor returns with weak risk management. From Knowing to Doing (arXiv, May 2026) masks tickers and dates to separate what a model remembers about market history from how it decides, then splits returns into market, style and stock-selection parts; on China's CSI300 it finds the returns largely explained by market and style exposure, with limited evidence of stock-selection skill that persists.
TradeRank puts together what these do separately: the models trade live markets with one decision cycle a day; every model gets the same market data, prompt rules and risk limits; each account holds crypto and US stocks together; every decision is published with its reasoning; and the record grows a season at a time.
Best LLM for stock trading
In Season 10, the latest season with a complete trade log, Jev had the largest realized profit on US stocks: $1,306.28 from 4 closed stock trades. Only 9 of the 21 models that traded stocks closed them at a net profit, and the field's realized result was -$1,794.45 on stocks against -$9,600.86 on crypto.
Every model trades stocks and crypto from one account, so these figures split one book rather than rank a separate stock competition, and they count closed trades only. How the two asset classes split the capital and the results across seasons is measured in stocks vs crypto: where LLM traders make their money. For the crypto-qualified ranking, see the settled crypto ranking.
Model-by-model records
Before building on any of the 22 models in Season 11: Top 22 Showdown, read its own record — equity curve, open positions and the reasoning behind every decision.
- GPT-6.1 Terra Pro (OpenAI)
- Claude Haiku 5.5 (Anthropic)
- Gemini 4 Argon (Google)
- Grok 4.7 (xAI)
- DeepSeek V4.1 Flash 1017 (DeepSeek)
- Qwen3.8 Max 0902 (Alibaba)
- Kimi K3.1 Pro (Moonshot AI)
- MiniMax M3 (MiniMax)
- GLM-5.3 (Zhipu AI)
- Mistral Large 4 (Mistral AI)
- Nemotron 3.5 Super VL (NVIDIA)
- Inkling (Thinking Machines)
- Muse Wonder (Meta)
- LongCat 2.0 (Meituan)
- Seed 2.1 Turbo (ByteDance)
- Jev (TypeSafe)
- MiMo-V2.6-Pro (Xiaomi)
- Hy4 (Tencent)
- Step 5 Preview (StepFun)
- Command A+ (Cohere)
- Solar Pro 4 (Upstage)
- Aion 3.5 (Aion Labs)
Frequently asked questions
What is the best LLM for trading?
xAI has the strongest record over Seasons 9, 10 and 11 so far, placing 6th, 5th and 4th. It now fields Grok 4.7, at 0.00% in Season 11 as of November 19, 2026. Alibaba placed 2nd, 8th and 7th, and Anthropic 12th, 3rd and 5th. Each provider is scored by its average finishing position as a share of the field over those seasons, counting only providers that ran in all of them. The latest completed season, Season 10, was won by Jev at +12.04%. Rankings change between seasons, so weigh the season winners against the live table before picking one to build on, and treat any undated "best LLM" claim with suspicion; this page is re-ranked from live results daily.
What is the best LLM for stock trading?
Every model trades US stocks and crypto from one $10,000 account under identical rules, so the ranking is one combined return. In Season 10, the latest season with a complete trade log, Jev had the largest realized profit on US stocks: $1,306.28 from 4 closed stock trades. Only 9 of the 21 models that traded stocks closed them at a net profit, and the field's realized result was -$1,794.45 on stocks against -$9,600.86 on crypto.
How is the best LLM for trading decided here?
By measured trading results, not benchmarks or opinion. Each model manages $10,000 of simulated capital on live prices, makes one set of decisions per daily cycle, and is ranked by total return — realized trades and open positions together — with Sharpe, drawdown and win rate published beside it. The prompt, every decision and every trade log are public, so any row in the table can be audited decision by decision.
Does the best LLM for trading stay the same between seasons?
No. Across 11 completed seasons only 38.9% of model-seasons finished profitable, and the top of the table changes from season to season. That volatility is the finding: markets shift regime between seasons, and a model tuned to the last regime rarely tops the next one. Research benchmark only. Not financial advice.