Model leaderboard
Every AI model, ranked
123 models rated on the benchmarks that matter, from paid work and coding agents to the hardest exams, like a chess rating: every benchmark two models share is a game between them. Open one for its price, its strengths and the source of every number.
Updated Nov 12, 2026.
- 1st
Claude Fable 5.22,275
- 2nd
Gemini 4 Carbon2,266
- 3rd
Claude Opus 5.52,045
- 4th
Gemini 4 Argon2,039
- 5th
GPT-6 Astra1,927
- 6th
Claude Fable 5.11,922
- 7th
Claude Sonnet 5.51,887
- 8th
GPT-6.1 Sol1,852
- 9th
Claude Opus 51,844
- 10th
Muse Wonder1,824
- 11th
GPT-6.1 Terra1,818
- 12th
GPT-5.5 Pro1,769
- 13th
Claude Fable 51,748
- 14th
Kimi K3.1 Pro1,722
- 15th
Muse Spark 1.31,700
- 16th
GPT-5.6 Sol1,677
- 17th
Gemini 3.7 Flash1,661
- 18th
Muse Ember1,654
- 19th
GPT-5.4 Pro1,600
- 20th
Gemini 3.8 Flash1,597
- 21st
GPT-6 Sol1,589
- 22nd
Qwen3.8 Max1,571
- 23rd
GLM-5.31,552
- 24th
Claude Opus 4.81,539
- 25th
Kimi K31,537
- 26th
Grok 4.61,532
- 27th
DeepSeek-V4.1-Flash1,530
- 28th
GPT-5.51,520
- 29th
Muse Spark 1.11,503
- 30th
Claude Opus 4.71,496
- 31st
GPT-5.6 Terra1,482
- 32nd
Seed 2.1 Pro1,476
- 33rd
GLM-5.3-Flash1,454
- 34th
Muse Spark 1.21,448
- 35th
Gemini 3.6 Flash1,440
- 36th
Qwen3.8-Flash-Next1,438
- 37th
DeepSeek-V4-Pro-08131,426
- 38th
Grok 4.51,421
- 39th
Grok 4.71,421
- 40th
Qwen3.7 Max1,418
- 41st
Claude Sonnet 51,413
- 42nd
Seed 2.1 Turbo1,403
- 43rd
Claude Haiku 5.51,386
- 44th
GLM-5.21,386
- 45th
Qwen3.8 Flash1,380
- 46th
GPT-5.41,379
- 47th
Gemini 3.5 Flash1,367
- 48th
Claude Opus 4.61,363
- 49th
Gemini 3.1 Pro1,353
- 50th
Gemini 3 Pro1,344
- 51st
Qwen3.8-27B1,306
- 52nd
Muse Spark Open1,305
- 53rd
Mistral Large 41,305
- 54th
Kimi K2.61,301
- 55th
GPT-5.6 Luna1,291
- 56th
Qwen3.6 Max1,277
- 57th
DeepSeek-V4-Pro-Max1,275
- 58th
GPT-6 Luna1,273
- 59th
Muse Spark1,249
- 60th
GPT-5.2 Pro1,249
- 61st
GLM-5.11,241
- 62nd
Claude Sonnet 4.61,222
- 63rd
Gemini 3 Flash1,211
- 64th
Mistral Large 41,197
- 65th
Qwen3.7-Plus1,194
- 66th
MiniMax M31,184
- 67th
Claude Opus 4.51,171
- 68th
Kimi K2.7 Code1,170
- 69th
DeepSeek-V4-Flash-07311,126
- 70th
GPT-5.21,125
- 71st
Qwen3.6 Plus1,118
- 72nd
DeepSeek-V4-Flash-Max1,117
- 73rd
GLM-51,115
- 74th
GPT-5.1 Thinking1,107
- 75th
GPT-51,038
- 76th
Kimi K2.51,020
- 77th
Nemotron 3.5 Super VL (120B A12B)999
- 78th
GLM-4.7994
- 79th
Qwen3.5-397B-A17B983
- 80th
Qwen3.5-122B-A10B978
- 81st
Claude Sonnet 4.5959
- 82nd
Qwen3.5-27B949
- 83rd
Muse Glimmer-30B943
- 84th
GPT-5.4 mini943
- 85th
MiniMax M2.5940
- 86th
Nemotron 3 Ultra (550B A55B)933
- 87th
Qwen3.6-27B920
- 88th
Gemini 3.5 Flash-Lite917
- 89th
Gemma 4 31B904
- 90th
GPT-5.5 Instant878
- 91st
Qwen3 Max Thinking862
- 92nd
Grok 4.3862
- 93rd
Gemini 2.5 Pro Preview 06-05858
- 94th
Qwen3.5-35B-A3B858
- 95th
Seed 1.8858
- 96th
Claude Opus 4.1853
- 97th
Gemini 3.1 Flash-Lite824
- 98th
DeepSeek-V3.2 (Thinking)799
- 99th
MiniMax M2.1788
- 100th
GLM-4.6777
- 101st
Qwen3.6-35B-A3B772
- 102nd
Gemma 4 26B-A4B765
- 103rd
DeepSeek-V3.2716
- 104th
GPT-5.4 nano709
- 105th
GPT-5 mini679
- 106th
Claude Haiku 4.5654
- 107th
Grok 4 Fast652
- 108th
Qwen3 VL 235B A22B Instruct636
- 109th
Seed 2.0 Mini633
- 110th
Qwen3.5 Flash612
- 111th
Claude Sonnet 4611
- 112th
Nemotron 3 Super (120B A12B)582
- 113th
GPT OSS 120B492
- 114th
DeepSeek-R1-0528485
- 115th
MiniMax M2473
- 116th
Kimi K2-Thinking-0905460
- 117th
Qwen3 VL 32B Instruct452
- 118th
Gemma 4 12B372
- 119th
Nemotron 3.5 Lightning (30B A3B)368
- 120th
Qwen3 VL 30B A3B Instruct353
- 121st
GPT-5 nano273
- 122nd
Qwen3 VL 8B Instruct171
- 123rd
Qwen3 VL 4B Instruct73