Pass IndexThe State of AISign in

Simplebench — Epoch AI

Simplebench — Epoch AI is run by Epoch AI. It has ranked 102 models, of which the catalogue holds 76, scoring from 0.107 to 0.819.

measured by Epoch AI · the board itself

PlaceModelMetricScore
1stof 102Claude Fable 5Score (AVG@5) (max)0.819
2ndof 102Claude Opus 5Score (AVG@5)0.806
3rdof 102Gemini 3.1 Pro PreviewScore (AVG@5)0.796
4thof 102GPT-5.5 ProScore (AVG@5)0.769
6thof 102Gemini 3.5 FlashScore (AVG@5)0.767
7thof 102gemini-3-pro-previewScore (AVG@5)0.764
8thof 102Grok 4.6Score (AVG@5)0.759
9thof 102Muse Spark 1.2Score (AVG@5)0.745
10thof 102GPT-5.4 ProScore (AVG@5)0.741
13thof 102Qwen3.7 MaxScore (AVG@5) (max)0.704
14thof 102Grok 4.5Score (AVG@5)0.7
15thof 102GPT-5.5Score (AVG@5)0.69
16thof 102Claude Opus 4.6Score (AVG@5)0.676
17thof 102Claude Opus 4.8Score (AVG@5)0.648
18thof 102GPT-5.6 SolScore (AVG@5)0.648
20thof 102Qwen3.6 Max PreviewScore (AVG@5) (max)0.63
21stof 102Gemini 2.5 Pro Preview 06-05Score (AVG@5)0.624
22ndof 102Claude Opus 4.5Score (AVG@5)0.62
24thof 102Claude Opus 4.7Score (AVG@5)0.617
25thof 102GPT-5 ProScore (AVG@5) (high)0.616
27thof 102Gemini 3 Flash PreviewScore (AVG@5)0.611
28thof 102DeepSeek V4 Flash (0731)Score (AVG@5)0.611
29thof 102Kimi K3Score (AVG@5) (max)0.607
30thof 102Claude Sonnet 5Score (AVG@5)0.606
32ndof 102Claude Opus 4.1Score (AVG@5)0.6
34thof 102Claude Opus 4Score (AVG@5)0.588
36thof 102GLM 5.2Score (AVG@5)0.588
37thof 102Kimi K2.7 CodeScore (AVG@5)0.579
38thof 102GPT-5.2 ProScore (AVG@5)0.574
39thof 102GPT-5Score (AVG@5) (high)0.567
40thof 102Grok 4.1 Fast Non-ReasoningScore (AVG@5)0.56
41stof 102Grok 4.1 Fast ReasoningScore (AVG@5)0.56
42ndof 102GLM 5.1Score (AVG@5)0.551
43rdof 102Claude Sonnet 4.5Score (AVG@5)0.543
45thof 102GPT-5.1Score (AVG@5) (high)0.532
46thof 102GLM-5Score (AVG@5)0.532
47thof 102o3Score (AVG@5) (high)0.531
48thof 102DeepSeek-V3.2-SpecialeScore (AVG@5)0.526
50thof 102Gemini 2.5 ProScore (AVG@5)0.516
51stof 102InklingScore (AVG@5)0.5
52ndof 102GPT-5.6 TerraScore (AVG@5)0.489
54thof 102GLM-4.7Score (AVG@5)0.477
56thof 102GPT-5.6 LunaScore (AVG@5)0.468
57thof 102Kimi K2.5Score (AVG@5)0.468
59thof 102Claude 3.7 SonnetScore (AVG@5)0.464
61stof 102DeepSeek V4 FlashScore (AVG@5)0.463
62ndof 102GPT-5.2Score (AVG@5) (high)0.458
64thof 102MiniMax M3Score (AVG@5)0.458
65thof 102Claude Sonnet 4Score (AVG@5)0.455
68thof 102o1Score (AVG@5)0.417
69thof 102Claude 3.5 SonnetScore (AVG@5)0.414
70thof 102Gemini 2.5 FlashScore (AVG@5)0.412
71stof 102DeepSeek-R1 (0528)Score (AVG@5)0.408
73rdof 102DeepSeek-V3.1Score (AVG@5)0.4
74thof 102o4-miniScore (AVG@5) (high)0.387
76thof 102Grok 3Score (AVG@5)0.361
77thof 102Qwen3.6-FlashScore (AVG@5)0.352
80thof 102Qwen3 235B A22BScore (AVG@5)0.31
81stof 102DeepSeek-R1Score (AVG@5)0.309
83rdof 102Llama-4-Maverick-17B-128E-InstructScore (AVG@5)0.277
85thof 102DeepSeek-V3 0324Score (AVG@5)0.272
87thof 102GPT-4.1 fine-tunedScore (AVG@5)0.27
88thof 102Kimi K2 InstructScore (AVG@5)0.263
89thof 102GPT-4 TurboScore (AVG@5)0.251
90thof 102Claude 3 OpusScore (AVG@5)0.235
91stof 102Meta-Llama-3.1-405B-InstructScore (AVG@5)0.23
92ndof 102o3-miniScore (AVG@5) (high)0.228
94thof 102Mistral Large 2407Score (AVG@5)0.225
95thof 102GPT OSS 120BScore (AVG@5)0.221
96thof 102Llama 3.3 70B InstructScore (AVG@5)0.199
97thof 102DeepSeek-V3Score (AVG@5)0.189
98thof 102Gemini 2.0 FlashScore (AVG@5)0.189
99thof 102o1-miniScore (AVG@5) (medium)0.181
100thof 102GPT-4o (2024-08-06)Score (AVG@5)0.178
101stof 102c4ai-command-r-plus-08-2024Score (AVG@5)0.174
102ndof 102GPT-4o-mini (2024-07-18)Score (AVG@5)0.107

Read from the board on 2026-09-16