Every model. Every benchmark. One table.

Curated catalog with lineage and sourced scores.

161 models · 7 benchmarks

run
columns
↗ ↗ ↗ ↗ ↗ ↗ ↗
Claude Mythos 5 81 $10 $50—94.1 —95.5 thinking 80.3 thinking 59 auto 88 max
Claude Opus 5 78 $5 $25—93.7 —96 79.2 56.3 52.6 90.8
Claude Opus 4.8 76 $5 $25—93.6 —88.6 69.2 49.8 84.3
Ornith-1.5-397B75 self-host self-host—92.8 —86 65.1 44.6 86.6
Claude Opus 5.575 $4 $20————89.9 max 64.4 max —
GPT-6 Astra74 $10 $50—96 max ————91.5 max
GPT-5.2 74 $1.75 $14—92.4 100 80 71.8 69 ———
Claude Mythos 5.174 — —————81.2 max 60.9 max —
Claude Sonnet 5.573 $2 $10————81.3 max 56.9 max —
Claude Opus 4.7 73 $5 $25—94.2 —87.6 thinking 64.3 thinking 46.9 max 79.3 max
Claude Fable 5.173 $10 $50————81.2 max 60.9 max —
GPT-5.6 73 $5 $30—94.6 94.1 ——64.6 —90.4
Kimi K372 self-host self-host—93.5 ———43.5 91.2
Claude Fable 5 72 $10 $50———95 thinking 80 thinking 53.3 —
Intern-S2-397B71 self-host self-host89.77 ———68.54 ——
Qwen3.8-Max (Qwen3.8-2.4T-A95B)69 $2 $6—92.6 ——67.7 43.6 —
Gemini 3.1 Pro 69 $2 $12—94.3 94.1 —80.6 54.2 44.4 85.9
Claude Sonnet 5 69 $3 $15———85.2 thinking 63.2 thinking 43.2 auto 84.7 max
Claude Opus 4.6 67 $5 $25—91.31 99.79 80.84 75.6 53.4 40 83.7
DeepSeek V4-Pro 66 $0.435 $0.8787.5 90.1 —80.6 55.4 37.7 83.4
Atria Dawn Preview65 self-host self-host————59.6 —92.5
Claude Haiku 5.565 $0.1 $0.5————64.8 max 45.9 max —
Hy365 self-host self-host—90.4 —78 57.9 53.2 —
GLM-5.2 64 $1.4 $4.4—91.2 ——62.1 40.5 —
Qwen3.5 397B A17B 63 $0.6 $3.687.8 88.4 —76.4 ——69
GPT-5.5 63 $5 $30—93.6 ——59.4 ——
GPT-5.4 63 $2.5 $15—92.8 —————
GPT-5.1 62 $1.25 $10—88.1 94 76.3 66 ———
MiniMax M3 62 $0.3 $1.2———80.5 59 —83.5
Claude Sonnet 4.6 62 $3 $15—89.9 95.6 79.6 58.1 34.6 74
Kimi K2.6 62 $0.95 $4—90.5 —80.2 58.6 36.4 83.2
Qwen3.8-Flash-Next61 $0.15 $0.47—91.7 ——62.5 35.9 —
Claude Opus 4.5 60 $5 $25—87 92.77 80.9 76.8 52 ——
Qwen3.6-27B59 self-host self-host86.2 thinking 87.8 thinking —77.2 thinking 53.5 thinking ——
GPT-5 59 $1.25 $10——94.6 74.9 65 ———
DeepSeek-V4.1-Flash59 $0.3 $1.2—90.9 ———36.8 —
Hy4 preview58 self-host self-host—————43.4 —
Kimi K2.5 58 $0.6 $387.1 thinking 87.6 thinking 96.1 thinking 76.8 70.8 50.7 31.5 60.6
DeepSeek V4-Flash 58 $0.14 $0.2886.2 88.1 —79 52.6 34.8 73.2
GPT-5.1-Codex-Max 58 $1.25 $10———77.9 ———
Gemma 4 31B Instruct57 self-host self-host85.2 84.3 —————
DeepSeek-V4-Pro-081357 $1.32 $3.96—————42.7 max —
Mistral Medium 3.557 $1.5 $7.5———77.6 ———
Qwen3.8-27B56 $0.5 $3—89.2 ——61.7 30.8 —
Ornith-1.5-35B-A3B56 self-host self-host—89.2 —79 59.6 25.6 67.6
Claude Sonnet 4.5 56 $3 $15———77.2 71.4 70.6 ———
DeepSeek-V4-Flash-073155 self-host self-host—————37.8 max —
Motif 354 self-host self-host—83.4 —76.2 —37 —
Gemini 3 Flash 53 $0.5 $3—90.4 —78 75.8 48.4 33.7 —
GLM-4.7 53 $0.6 $2.284.3 85.7 95.7 73.8 —24.8 52
Gemma 4 26B A4B Instruct52 self-host self-host82.6 82.3 —————
Gemini 3.6 Flash 52 $1.5 $7.5————58.7 ——
Gemini 3.5 Flash 52 $1.5 $9————53.9 40.2 —
GLM-5 51 $1 $3.2—86 —77.8 72.8 —30.5 62
Gemini 3 Pro 51 — ——91.9 —76.2 69.6 43.3 37.5 59.2
GLM-5.1 50 $1.4 $4.4—86.2 ——58.4 31 68
MiniMax M2.1 49 $0.3 $1.288 83 83 74 —22.2 47.4
Devstral 248 $0.4 $2———72.2 ———
DeepSeek R1-0528 48 self-host self-host85 81 87.5 57.6 ———
A.X K248 self-host self-host—85.6 thinking ———27.8 thinking —
Gemini 3.5 Flash-Lite 48 $0.3 $2.5————54.2 ——
Qwen3.6-35B-A3B48 self-host self-host85.2 thinking 86 thinking —73.4 thinking 49.5 thinking 21.4 thinking —
Nemotron 3 Ultra 550B A55B 48 self-host self-host86.8 ——70.7 —26.7 44.4
Ministral 3 14B Reasoning 47 $0.2 $0.2——85 ————
GLM-4.547 $0.6 $2.284.6 79.1 —64.2 54.2 ———
Muse Glimmer 30B47 self-host self-host———76 high 51.2 high ——
DeepSeek V3.2-Speciale 47 self-host self-host—————30.6 —
Ornith-1.0-35B45 self-host self-host———75.6 50.4 ——
DeepSeek V3.2 45 self-host self-host———73.1 70 —25.1 —
Devstral Small 244 $0.1 $0.3———68 ———
Grok 344 — —79.9 75.4 —————
K-EXAONE 2.043 self-host self-host83.5 reasoning 82.2 reasoning —68.2 —18.3 reasoning —
Gemini 2.5 Pro 43 $1.25 $10—86.4 88 59.6 53.6 —21.6 —
DeepSeek V3.2-Exp 43 self-host self-host85 79.9 89.3 67.8 —19.8 40.1
Kolibri 143 self-host self-host80 reasoning 84.3 reasoning 96.9 reasoning 66.4 —21.5 reasoning 29.4
DeepSeek R1 43 self-host self-host—71.5 —————
Claude Haiku 4.5 43 $1 $5—73 80.7 73.3 66.6 ———
Llama 4 Maverick Instruct42 self-host self-host80.5 69.8 —21 ———
Gemma 4 12B Instruct42 self-host self-host77.2 78.8 —————
Claude Opus 4 41 $15 $75—79.6 thinking 75.5 thinking 72.5 none 67.6 ———
Qwen3 235B A22B 41 $0.7 $2.8—71.1 81.5 ————
Mistral Large 340 $0.5 $1.5—67.17 —————
Qwen3-Coder-Next39 self-host self-host———70.6 44.3 ——
Nemotron 3 Super 120B A12B 39 self-host self-host83.73 —90.21 60.47 —18.26 31.28
Ornith-1.5-9B39 self-host self-host—86.4 —70.6 47.5 20.2 56.4
DeepSeek R1-Distill-Qwen-32B 39 self-host self-host—62.1 —————
Claude 3.5 Sonnet38 — ——59.4 —————
Ornith-1.0-9B38 self-host self-host———69.4 42.9 ——
Claude Sonnet 4 38 $3 $15—75.4 thinking 70.5 thinking 72.7 none 64.9 ———
Olmo 3-Think 32B 37 self-host self-host——72.5 ————
Gemini 2.0 Flash37 — —77.6 60.1 —13.5 ———
Granite 4.2 30B37 self-host self-host77.6 thinking —89.17 thinking 57 33.29 ——
Gemini 3.1 Flash-Lite 37 $0.25 $1.5—86.9 ——38.3 16 —
Devstral Small 1.137 self-host self-host———53.6 ———
Llama 3.1 405B Instruct37 self-host self-host—51.1 —————
Claude 3.7 Sonnet 36 — ——68 —63.7 52.8 ———
Granite 4.2 3B36 self-host self-host67.84 thinking —78.33 thinking ————
MiniMax M1 36 self-host self-host81.1 70 76.9 56 ———
GLM-4.7-Flash 36 self-host self-host——91.6 59.2 —14.4 42.8
DeepSeek R1 Distill Llama 8B 35 self-host self-host—49 —————
MiniMax M2 35 $0.3 $1.282 78 78 69.4 61 —12.5 44
o1 35 $15 $60—78.3 —40.9 ———
DeepSeek V3.1 34 self-host self-host—80.1 88.4 66 —15.9 30
Kimi K234 self-host self-host81.1 75.1 49.5 65.8 ———
LFM2.5-2.6B34 self-host self-host——51.87 thinking ————
Qwen3 32B 34 $0.16 $0.64—68.4 72.9 ————
Llama 4 Scout Instruct34 self-host self-host74.3 57.2 —9.1 ———
GLM-4.634 $0.6 $2.2———68 55.4 —17.2 45.1
Nemotron 3 Nano 30B A3B 33 self-host self-host78.3 —89.1 38.8 —10.6 —
Magistral Medium 33 $2 $5—70.8 64.9 ————
GPT-4.133 $2 $8—66.3 —54.6 39.6 ———
FrogNano-4B33 self-host self-host———61.5 37.6 ——
Qwen2.5 72B Instruct32 self-host self-host71.1 49 —————
Llama 3.3 70B Instruct31 self-host self-host68.9 50.5 —————
Granite 4.2 8B30 self-host self-host74.04 thinking —86.67 thinking 47.67 19.11 ——
NVIDIA Nemotron 3.5 Lightning 30B-A3B30 self-host self-host81.94 thinking 75.44 thinking —51.56 —11.72 thinking 36.97
Gemini 2.5 Flash 30 $0.3 $2.5—80.8 75.6 54 28.7 —11 —
DeepSeek V329 self-host self-host75.9 59.1 —42 ———
Gemma 3 27B Instruct28 self-host self-host67.5 42.4 —————
Llama 3.1 70B Instruct28 self-host self-host66.4 48 —————
MiniCPM5-2B28 self-host self-host70.8 —86.5 46.4 14.4 ——
Granite 4.1 30B27 self-host self-host64.09 45.76 —————
Qwen2.5-Coder 32B Instruct26 self-host self-host62.3 41.8 —9 ———
Mellum2.1 Thinking25 self-host self-host—64.6 thinking —47 thinking 28 thinking ——
Llama 3.1 8B Instruct25 self-host self-host48.3 31.8 —————
Gemini 1.5 Pro—— —65.7 37.1 ———3.9 —
Gemini 3.7 Flash—— ————————
Gemini 3.8 Flash—— ————————
Gemini 3 Deep Think —— ————————
Gemini 4 Argon—— ————————
Gemma 2 27B—self-host self-host———————
GLM-5.3—$1.4 $4.4———————
GLM-5.3-Flash—$0.15 $0.5———————
GPT-4o—$2.5 $1074 52.6 —21.6 —2.8 —
GPT-5.2-Codex —$1.75 $14—89.9 ———33.5 —
GPT-5.3-Codex —$1.75 $14—91.5 ———39.9 —
GPT-6.1 Sol—$2 $10———————
GPT-6 Luna—$0.1 $0.5———————
GPT-6 Sol—$2 $10———————
Grok 4 —— —86.6 87.7 92.7 ——23.9 —
Grok 4.1 —— ————————
Grok 4.1 Fast—— —85.4 74.3 85.3 63.7 89.3 34.3 ——17.6 5 —
Grok 4.5 —$2 $6—93.1 ———40.3 —
Grok 4.6—$2 $6———————
Grok 4.7—$2 $6———————
Hermes 3 Llama 3.1 8B—self-host self-host———————
Kimi K2.7-Code —$0.95 $4—89.6 ———32.8 —
Ling-3.0-flash—self-host self-host———————
Ling-3.0-tiny—self-host self-host———————
Llama 3.1 405B—self-host self-host73.2 51.5 3 ——4.2 —
Llama 3.1 8B—self-host self-host47.6 25.9 4.3 ——5.1 —
Llama 3.2 3B Instruct—self-host self-host34.7 22.4 3.3 ——5.2 —
MiMo-V2.6-Flash—$0.14 $0.28———————
MiMo-V2.6-Pro—$0.435 $0.87———————
Mistral Large 2—self-host self-host68.3 47.2 0 ——3.2 —
Mistral Large 4—$1.36 $4.18———————
Mixtral 8x22B—$2 $653.7 33.2 ———4.1 —
Muse Spark 1.2—$1.25 $4.25———————
Muse Spark 1.3—$1.25 $4.25———————
o3 —$2 $885.3 82.7 88.3 58.4 —20 —
o4-mini —$1.1 $4.483.2 78.4 90.7 45 —17.5 —

how to read this

  • Every figure is published by the model's maker. We run no benchmarks of our own. Each score links to the model card, paper or announcement it was taken from — follow it before you rely on it.
  • An empty cell means unpublished, not zero. Makers report the benchmarks that suit them, so a sparse row is a claim not made rather than a claim failed.
  • Overall ranks each model within every column it enters, then averages those standings and pulls thin evidence toward the middle: one reported benchmark is a weaker claim than five, and the number says so.
  • A figure in this colour was measured by someone else. Independent platforms run the evaluation themselves, the same way for every model, and publish what they got. Where one sits beside a maker's own number, the gap between them is the interesting part; where it sits alone, it is the only measurement anyone has published. Neither is automatically the truer figure — but the ranking is built only from what makers publish, so a measured figure never moves a model up the table.
  • A score is also a claim about effort. The same model answers differently at max reasoning and at its default setting, and makers headline whichever suits them. Where the source states the setting it is marked on the figure; where it does not, the tooltip says so rather than pretending the run was standard.
  • Compare down a column, not across the table. Makers evaluate on their own harnesses, and two labs reporting the same benchmark have not necessarily run the same test.