Research benchmark only. Not financial advice. Crypto trading can lose all capital.
AI Trading Ethereum: 37 LLMs, 425 Trades, 6 Seasons
Ethereum has been tradeable since the arena opened, so its record runs the full length of the archive. Most competitors have held ether at some point, argued with themselves about it in their reasoning, and eventually written down why they let go.
Kimi K3 holds the strongest aggregate realized result on Ethereum: $699.68 across 4 closed trades. That is one asset's record, not an overall ranking — 37 models have traded an ETH position here, over 6 seasons whose trade log survives.
- 425ETH trades on record (floor)
- 203closed with a realized result
- -$940.28realized, all models combined
- 7open positions in the live season
Best LLM for Trading Ethereum
The leading ETH results rest on a handful of closed legs each, and the biggest gains among them came late in the archive. Further down the full table, the first row from the early seasons closed many more legs for a smaller total. Same column, same units, very different amounts of evidence, which is why every entry states how many legs it closed.
Kimi K3 — $699.68 realized on ETH
4 closed legs of 8 trades, in Seasons 8 and 9.
MiniMax M3 — $636.12 realized on ETH
5 closed legs of 13 trades, in Seasons 6, 8 and 9.
Inkling — $332.00 realized on ETH
2 closed legs of 4 trades, in Seasons 8 and 9.
Ranked by aggregate realized profit and loss on ETH alone, over the seasons whose trade log survives. Equal results are joint, never ordered. Open positions carry no realized figure and are excluded.
Can AI Trade Ethereum?
They already do, and an ETH position is ordinary business: opened on a stated thesis, sized as a share of equity, closed when the thesis breaks or the season ends. What varies is the justification, and that is quoted in full further down.
37 model versions have executed an ETH trade in TradeRank, and 50 versions have traded it or written about it in their decisions. The realized total across every closed leg is -$940.28, and that total is a floor rather than a count.
Every Model's ETH Record
One row per model version. A cell reads n/a where the answer is unknown: the trade log for those seasons did not survive. A known zero is shown as a zero instead — a row that traded and closed nothing realized nothing, which is a measurement rather than a gap.
| Model | Provider | Trades | Closed | Realized P&L | Seasons |
|---|---|---|---|---|---|
| Kimi K3 | Moonshot | 8 | 4 | $699.68 | Seasons 8 and 9 |
| MiniMax M3 | MiniMax | 13 | 5 | $636.12 | Seasons 6, 8 and 9 |
| Inkling | Thinking Machines | 4 | 2 | $332.00 | Seasons 8 and 9 |
| Gemini 3 Pro | 24 | 12 | $265.88 | Seasons 1 and 2 | |
| Claude Fable 5 | Anthropic | 5 | 3 | $258.74 | Seasons 6 and 8 |
| Nemotron 3 Ultra | NVIDIA | 7 | 3 | $220.64 | Seasons 6, 8 and 9 |
| ox-alpha | Stealth | 3 | 1 | $77.35 | Season 8 |
| Mistral Medium 3.5 | Mistral AI | 4 | 2 | $68.78 | Seasons 6 and 8 |
| GLM-5.3 | Zhipu AI | 2 | 1 | $59.15 | Season 9 |
| Grok 4-1 Fast | xAI | 37 | 20 | $47.52 | Seasons 0 and 1 |
| Qwen3.8 2.4T A95B | Alibaba | 3 | 1 | $39.12 | Season 8 |
| Claude Sonnet 4.5 | Anthropic | 2 | 1 | $12.19 | Season 1 |
| Claude Opus 4.8 | Anthropic | 2 | 0 | $0.00 | Season 6 |
| GLM-5.2 | Zhipu AI | 4 | 2 | -$23.13 | Season 6 |
| GPT-5.5 | OpenAI | 2 | 1 | -$24.01 | Season 6 |
| Kimi K2.7 Code | Moonshot | 4 | 2 | -$26.92 | Season 6 |
| Seed 2.1 Turbo | ByteDance | 2 | 1 | -$30.68 | Season 9 |
| Qwen 3.7 Plus | Alibaba | 2 | 1 | -$35.70 | Season 6 |
| Claude Fable 5.1 | Anthropic | 2 | 1 | -$38.35 | Season 9 |
| Qwen3.8 Max 0902 | Alibaba | 2 | 1 | -$38.35 | Season 9 |
| Muse Spark 1.3 | Meta | 2 | 1 | -$46.02 | Season 9 |
| Claude Haiku 4.5 | Anthropic | 56 | 26 | -$70.28 | Seasons 0 and 1 |
| Gemini 3.5 Flash | 2 | 1 | -$72.97 | Season 6 | |
| Gemini 2.5 Flash | 20 | 10 | -$80.60 | Seasons 0 and 1 | |
| DeepSeek V4 Pro | DeepSeek | 6 | 2 | -$83.13 | Season 6 |
| Nightjar (Hy4 Preview) | Stealth | 2 | 1 | -$92.05 | Season 9 |
| Grok 4.3 | xAI | 2 | 1 | -$126.74 | Season 6 |
| o4-mini | OpenAI | 20 | 10 | -$126.90 | Season 1 |
| Nemotron 3.5 Lightning | NVIDIA | 1 | 1 | -$150.10 | Season 9 |
| DeepSeek V4.1 Flash | DeepSeek | 4 | 2 | -$212.45 | Season 9 |
| MiMo-V2.6-Pro | Xiaomi | 2 | 1 | -$213.75 | Season 9 |
| Qwen3 | Alibaba | 47 | 24 | -$215.63 | Season 1 |
| GPT-5 Mini | OpenAI | 45 | 18 | -$247.53 | Seasons 0 and 1 |
| LongCat 2.0 | Meituan | 6 | 3 | -$249.45 | Season 9 |
| Gemini 3.8 Flash | 4 | 2 | -$269.98 | Season 9 | |
| Kimi K2 | Moonshot | 30 | 15 | -$581.34 | Season 1 |
| DeepSeek R1 | DeepSeek | 44 | 21 | -$601.39 | Season 1 |
| Claude Opus 4.6 | Anthropic | n/a | n/a | n/a | Season 3 |
| Claude Opus 4.7 | Anthropic | n/a | n/a | n/a | Seasons 4 and 5 |
| DeepSeek V3.2 | DeepSeek | n/a | n/a | n/a | Season 3 |
| Gemini 3.1 Pro | n/a | n/a | n/a | Seasons 3 and 4 | |
| GLM-5.1 | Zhipu AI | n/a | n/a | n/a | Seasons 4 and 5 |
| Grok 4.20 | xAI | n/a | n/a | n/a | Season 3 |
| Grok 4.20 MA | xAI | n/a | n/a | n/a | Season 4 |
| Kimi K2.5 | Moonshot | n/a | n/a | n/a | Season 3 |
| Kimi K2.6 | Moonshot | n/a | n/a | n/a | Seasons 4 and 5 |
| MiniMax M2.5 | MiniMax | n/a | n/a | n/a | Season 3 |
| MiniMax M2.7 | MiniMax | n/a | n/a | n/a | Season 4 |
| Qwen 3.5 Plus | Alibaba | n/a | n/a | n/a | Season 3 |
| Qwen 3.6 Plus | Alibaba | n/a | n/a | n/a | Seasons 4 and 5 |
Ethereum AI Trading Performance by Season
Ethereum concentrates most of its executed volume in one early season and then thins out sharply. The middle of the archive kept models writing about ether every cycle without leaving a single recoverable fill behind.
Trade counts are not comparable across the whole column: the early seasons ran several decision cycles a day and the later ones settle everything at one fixed hour, so the Cycle column states which regime each row belongs to.
| Season | Cycle | Trades | Closed | Realized P&L | Models trading |
|---|---|---|---|---|---|
| Season 0 | intraday | 28 | 15 | $95.27 | 4 |
| Season 1 | intraday | 295 | 141 | -$1,687.72 | 10 |
| Season 2 | intraday | 2 | 1 | -$5.63 | 1 |
| Season 3 | once daily | n/a | n/a | n/a | n/a |
| Season 4 | once daily | n/a | n/a | n/a | n/a |
| Season 5 | once daily | n/a | n/a | n/a | n/a |
| Season 6 | once daily | 37 | 17 | -$766.76 | 12 |
| Season 7 | once daily | n/a | n/a | n/a | n/a |
| Season 8 | once daily | 23 | 10 | $2,955.51 | 8 |
| Season 9 | once daily | 40 | 19 | -$1,530.96 | 15 |
A cell reads n/a where the answer is unknown: the trade log for those seasons did not survive. A known zero is shown as a zero instead — a row that traded and closed nothing realized nothing, which is a measurement rather than a gap.
Who Holds ETH Right Now
The season now running, as of 2026-10-10. Every table above this block is frozen archive; this block and the fourth figure in the hero are the only parts that move.
| Model | Side | Entry | Last price |
|---|---|---|---|
| Aion 3.5 | long | $2,499.28 | $2,498.45 |
| DeepSeek V4.1 Flash | long | $2,499.28 | $2,498.45 |
| Inkling | long | $2,499.28 | $2,498.45 |
| Kimi K3 | long | $2,499.28 | $2,498.45 |
| Mistral Large 4 | long | $2,499.28 | $2,498.45 |
| Seed 2.1 Turbo | long | $2,499.28 | $2,498.45 |
| Step 5 Preview | long | $2,499.28 | $2,498.45 |
Season so far: 7 trades, 0 closed legs, realized n/a. The live leaderboard carries the current standings.
Entering a long position in ETH at a short-term oversold level within a confirmed medium-term uptrend offers attractive risk-reward for a multi-week holding period.
Building a 40% long in ETH provides concentrated, liquid exposure to crypto recovery with a structurally-defined invalidation near recent support, consistent with a patient, medium-term holding period.
Buying a well-defined support zone in an intact weekly uptrend after an overleveraged flush offers ~4% defined risk against ~11% reward over a multi-week horizon.
The Best and Worst ETH Calls
Selected deterministically from the trade log: the strongest and weakest closed ETH legs that carry a written rationale, whichever model made them. A larger leg with no reasoning attached to it is not shown here, because the section exists to publish the reasoning.
Largest quoted ETH leg: $1,124.22, MiniMax M3
Original target of $2,200 was reached; now at $2,326 with 4h RSI 94.62 signaling extreme overbought. Full exit locks in +22.6% gain and frees capital to redeploy at better risk/reward after a pullback to retest breakout…
Weakest quoted ETH leg: -$194.99, MiniMax M3
The bull flag above $2,650 — the actual trade structure that anchored the thesis — has broken on heavy volume with funding flipping negative; preserving capital at -2.2% is materially better than holding a…
What the Models Said About Ethereum
Every quote below is a model's own reasoning for an ETH decision, unedited apart from length. Selection is deterministic and recorded in the pack.
The bull flag above $2,650 — the actual trade structure that anchored the thesis — has broken on heavy volume with funding flipping negative; preserving capital at -2.2% is materially better than holding a…
The breakout above $2,550 has failed with price breaking down back into the prior range and breaching the $2,430 invalidation level.
The thesis is invalidated by my own rule: ETH touched 2400 (low 2389.7), signaling the multi-week breakout range failed, so I exit rather than average down.
The bull flag thesis is invalidated: price broke below the $2,405–$2,537 consolidation range and the $2,387 support, with the 4h chart showing sustained lower highs ($2,515 → $2,398 → $2,422 → $2,393) and no recovery.…
Opening a high-conviction 60% core long in ETH because it combines the strongest trend, proximity to highs, and deepest liquidity among tradeable crypto, with cash retained for future adds.
Opening a 50% long in ETH to capture the resumption of the medium-term uptrend after a healthy correction, using a hard invalidation just below the structural low.
Enter the strongest major while structure is intact, risking ~7.5% on the position (~2.2% of equity) against a multi-week continuation setup.
BTC breakout provides tailwind for ETH with less overbought risk than other momentum names and clean technical structure for a medium-term move to 3000.
ETH is the strongest liquid crypto benchmark trade available today while equities are closed.
Strongest major pressing range highs with flat funding — moderate starter size with room to add on confirmation.
Best risk/reward in crypto universe: strong trend, good momentum, not overextended, deep liquidity.
ETH has the strongest uptrend structure among large-cap crypto assets, high liquidity, and favorable risk-reward for medium-term trend continuation.
Tightest and most constructive medium-term structure among tradeable assets today; 20% sizing caps weekend risk while preserving capital for equity ideas when US markets reopen Monday.
Holding above 2400 support after crash to 2406 and bounce to 2488, weekly breakout structure intact
ETH offers diversification from SOL with complementary technical structure and sufficient distance from existing positions' invalidation levels
Original target of $2,200 was reached; now at $2,326 with 4h RSI 94.62 signaling extreme overbought. Full exit locks in +22.6% gain and frees capital to redeploy at better risk/reward after a pullback to retest breakout…
ETH reached its rebound target with sharply overbought RSI, so I am locking in profits per the thesis-completion rule.
Banking half the +23% gain with price well through the 2,050-2,180 thesis target; keeping a runner for the neutral-funding regime shift.
Original oversold bounce thesis invalidated by price action: ETH weekly RSI now above 50 (53.11) and 4h RSI overbought (87.88). Locking in +10.8% profit.
Target $2100-2200 reached; weekly RSI normalized from oversold; mean-reversion thesis complete.
Buy the confirmed range breakout in the highest-liquidity crypto with uncrowded positioning and room on the weekly.
A small position adds exposure to the strongest tradeable crypto setup without overcommitting to an overbought tape.
Deploying the previously telegraphed ETH trigger — a healthy cool-off that held the breakout base — at measured size with the invalidation placed where the flag is objectively dead.
ETH offers the strongest technical improvement among liquid tradeables, allowing a patient, well-defined long with clear invalidation criteria.
What This Page Does and Does Not Measure
- Trade records cover Seasons 0, 1, 2, 6, 8 and 9. Seasons 3, 4, 5 and 7 ran Ethereum but their trade logs are unrecoverable, so those rows read n/a.
- Season 1 survives only as a partial log, so every all-time total on this page is a floor rather than a count.
- Decision reasoning covers Seasons 1, 3, 4, 5, 6, 7, 8 and 9, which is why a model can be quoted in a season with no trades beside it.
- 32.0% of the 203 closed ETH legs finished positive. This is a closed-trade rate on this asset only. It is not the standings win rate, which counts open positions too, and the two are not comparable in either direction.
- Everything here is paper trading. Each model starts from the same simulated balance, pays a fee on every fill, and risks no capital. How the competition works.
Ethereum AI Trading — FAQ
What is the best LLM for trading Ethereum?
Kimi K3 holds the strongest aggregate realized result on ETH: $699.68 across 4 closed trades. That is one asset's record, not an overall ranking, and it covers only the seasons whose trade log survives.
Can AI trade Ethereum?
They already do. 37 models have executed 425 ETH trades in TradeRank, 203 of them closed with a realized result.
How many AI models have traded ETH?
37 model versions have executed an ETH trade. A wider set — 50 versions — traded it or wrote about it in their decisions, including seasons whose trade log did not survive.
Did the models make or lose money on Ethereum?
Across the closed ETH trades on record the realized total is -$940.28. Season 1 survives only as a partial log, so that is a floor. Open positions carry no realized figure and are not in that total.
Which seasons does this Ethereum record cover?
Trades: Seasons 0, 1, 2, 6, 8 and 9. Seasons 3, 4, 5 and 7 ran the asset but their trade logs are unrecoverable, so those rows read n/a rather than zero.
Are the Ethereum results on this page live trades?
No. TradeRank is a paper-trading benchmark: each model starts from the same simulated balance, pays fees on every fill, and risks no capital.