Enkidu · model inventory
Which coding model is good at what, and how fast.
Every model Enkidu confirmed it can run, rated per skill from published leaderboards (SWE-bench, Terminal-Bench, Design Arena, OpenRouter τ²-bench), ranked against every other model on the same board. Speed is the time to finish the same small coding task.
yellow = approximate: no result on our boards, so it takes the profile of the closest rated model from an announcement page's comparison table~ = one leaderboard onlyestimate from search= no benchmark number exists, so the standing is read off what search results say the model is (the quoted page); the weakest kind of number herefound by Arca research= a published figure Arca found on the web and verified on the cited pageupdated 2026-09-25 15:25 UTC · 25 models, 1 approximate
| model | agent | speed | overall | repo | terminal | backend | fullstack | frontend | algo | tools | game | 3D | data viz | sci | thinking | price |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
claude-fable-5 leaderboardfound by Arca research | claude anthropic | ▮▮▯▯▯ 11.3s | ▮▮▮▮▮ 90 | 85 | 71 | 97 | 87 | 92~ | ≈100 | 99~ | 96~ | 92~ | 92~ | · | low → max | included in your claude plan |
fable → claude-fable-5-1 leaderboardfound by Arca research | claude anthropic | ▮▮▮▯▯ 9.2s | ▮▮▮▮▮ 89 | 77 | 53 | 96 | 97 | 97~ | 82~ | 96~ | 99~ | 99~ | 99~ | · | low → max | included in your claude plan |
gpt-6-astra leaderboardfound by Arca research | codex openai | ▮▯▯▯▯ 17.1s | ▮▮▮▮▮ 86 | 60 | 54 | · | 100~ | 95~ | · | · | 100~ | 100~ | 95~ | · | low → ultra · default medium | included in your codex plan |
opus → claude-opus-5-5 leaderboardfound by Arca research | claude anthropic | ▮▮▮▮▯ 8.4s | ▮▮▮▮▮ 82 | 81 | 80~ | · | ≈100 | 100~ | · | 67~ | ≈100 | ≈100 | ≈100 | · | low → max | included in your claude plan |
gpt-5.6-sol leaderboardfound by Arca research | codex openai | ▮▮▮▮▮ 7.3s | ▮▮▮▮▮ 80 | 62 | 48 | 75~ | 81~ | 97~ | ≈61 | 75 | 98~ | 97~ | 94~ | · | low → ultra · default low | included in your codex plan |
gpt-6-sol leaderboardfound by Arca research | codex openai | ▮▮▯▯▯ 13.5s | ▮▮▮▮▯ 78~ | 76~ | · | · | · | 80~ | · | 46~ | 92~ | · | 100~ | · | low → ultra · default medium | included in your codex plan |
gemini-3.7-flash leaderboardfound by Arca research | opencode google | ▮▮▮▮▯ 8.1s | ▮▮▮▮▯ 75 | 40 | 43 | · | 74 | 93~ | ≈48 | ≈47 | 94~ | 92~ | 93~ | · | low → high | $0.75 in · $3.75 out / M tok |
gemini-3.8-flash leaderboardfound by Arca research | opencode google | ▮▯▯▯▯ 22.9s | ▮▮▮▮▯ 75 | 49 | 52 | · | 74~ | 94~ | ≈52 | ≈52 | 93~ | 90~ | 79~ | · | low → high | $0.75 in · $3.75 out / M tok |
sonnet → claude-sonnet-5 leaderboardfound by Arca research | claude anthropic | ▮▮▮▮▮ 6.3s | ▮▮▮▮▯ 75 | 55 | 42 | 92~ | 79 | 86~ | · | 79~ | 89~ | 84~ | 74~ | · | low → max | included in your claude plan |
gemini-flash-latest ≈ Arca's model identified it as Gemini 3.8 Flash (medium confidence) from https://openrouter.ai/google — "Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning." estimate from search · medium confidenceopenrouter.ai: “Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.” | opencode google | ▮▮▮▮▮ 3.6s | ≈ 75 | ≈49 | ≈52 | · | ≈74 | ≈94 | · | · | ≈93 | ≈90 | ≈79 | · | low → high | $0.75 in · $3.75 out / M tok |
gemini-3.6-flash leaderboardfound by Arca research | opencode google | ▮▮▮▯▯ 10.2s | ▮▮▮▮▯ 73 | 63 | 66 | 62 | 52 | 91~ | ≈97 | 62~ | 82~ | 87~ | 92~ | · | minimal → high | $0.75 in · $3.75 out / M tok |
muse-spark-1.3-contributor-free leaderboardfound by Arca research | opencode opencode | ▮▮▮▯▯ 9.2s | ▮▮▮▮▯ 73 | 40 | 66 | 12~ | 79 | 99~ | · | ≈75 | 97~ | 99~ | 96~ | · | minimal → xhigh | free |
gpt-5.5 leaderboardfound by Arca research | codex openai | ▮▮▮▮▮ 7.1s | ▮▮▮▮▯ 68 | 72 | 88 | 51~ | 21~ | 75~ | ≈65 | 75 | 91~ | 66~ | 77~ | · | low → xhigh · default medium | included in your codex plan |
gpt-6-luna leaderboardfound by Arca research | codex openai | ▮▮▮▯▯ 9.0s | ▮▮▮▮▯ 63~ | 45~ | ≈55 | ≈66 | · | 82~ | · | 35~ | 87~ | ≈42 | 70~ | · | low → max · default medium | included in your codex plan |
gemini-3.5-flash leaderboardfound by Arca research | opencode google | ▮▮▮▮▯ 7.5s | ▮▮▮▮▯ 62 | 41 | 55 | 36 | 51 | 80~ | · | 73~ | 84~ | 79~ | 67~ | · | minimal → high | $1.5 in · $9 out / M tok |
gpt-5.6-luna leaderboardfound by Arca research | codex openai | ▮▮▯▯▯ 15.8s | ▮▮▮▮▯ 61 | 68 | 58~ | 69~ | ≈17 | 76~ | ≈55 | 55 | 65~ | 44~ | 59~ | · | low → max · default medium | included in your codex plan |
gpt-5.6-terra leaderboardfound by Arca research | codex openai | ▮▮▮▮▯ 7.7s | ▮▮▮▯▯ 55 | 51 | 44 | ≈73 | 17 | 79~ | 56~ | 55 | 74~ | 49~ | 73~ | · | low → ultra · default medium | included in your codex plan |
gemini-3-flash-preview leaderboardfound by Arca research | opencode google | ▮▮▯▯▯ 11.5s | ▮▮▮▯▯ 53 | 45 | 29 | 58~ | 27 | 60~ | · | 94~ | 52~ | 62~ | · | · | minimal → high | $0.5 in · $3 out / M tok |
gemini-3.1-pro-preview leaderboardfound by Arca research | opencode google | ▮▮▯▯▯ 14.4s | ▮▮▮▯▯ 51 | 16 | 21 | · | 28~ | 68~ | · | 76~ | 66~ | 75~ | 58~ | · | low → high | $2 in · $12 out / M tok |
mimo-v2.6-flash-free found by Arca research | opencode opencode | ▮▮▮▮▯ 8.2s | ▮▮▮▯▯ 48~ | 31~ | 31~ | · | · | · | · | 84~ | · | · | · | · | yes | free |
big-pickle found by Arca research · the page hedges the figure (~) | opencode opencode | ▮▯▯▯▯ 24.6s | ▮▮▮▯▯ 47~ | 47~ | 47~ | 47~ | 47~ | · | · | · | · | · | · | · | yes | free |
nemotron-3-ultra-free leaderboardfound by Arca research | opencode opencode | ▮▯▯▯▯ 19.4s | ▮▮▮▯▯ 43 | 41 | 28 | 45~ | 45~ | 37~ | · | 78 | 43~ | 48~ | 30~ | · | yes | free |
space-bunny-free found by Arca research | opencode opencode | ▮▮▮▯▯ 10.5s | ▮▮▯▯▯ 29~ | 29~ | 29~ | 29~ | 29~ | ≈82 | ≈70 | ≈82 | ≈84 | ≈84 | ≈84 | · | low → max | free |
nemotron-3.5-lightning-free leaderboardfound by Arca research | opencode opencode | ▮▯▯▯▯ 18.1s | ▮▯▯▯▯ 18~ | 2~ | · | · | · | · | · | 34~ | · | · | · | · | yes | free |
gemini-2.5-flash leaderboardfound by Arca research | opencode google | ▮▮▮▮▮ 6.6s | ▮▯▯▯▯ 15 | 15~ | · | · | · | 19~ | 15~ | 21 | 16~ | 14~ | 11~ | · | yes | $0.3 in · $2.5 out / M tok |
Index = the model's best published figure on each board, ranked 0–100 against every model on that board, weighted per skill. Hover a cell for the boards behind it. Rows come from the Enkidu inventory on one machine and refresh with it; a ≈ row is a stand-in until the model earns a row on a board we read.