Mock AIME 2024–25
Competition mathematics, from the OTIS Mock AIME 2024-2025 set. Aggregated from Epoch AI's AI Benchmarking Hub (otis_mock_aime_2024_2025.csv, column 'mean_score'), used under CC-BY 4.0. The Known Good aggregates published results; it runs no evaluations.
238 models scored
Higher is better
Published by Epoch AI
CC-BY 4.0
Mock AIME 2024–25
Mock AIME 2024–25 as measured and published by Epoch AI (CC-BY 4.0). Ingested, normalised only where it feeds an index, and never re-run by us. Higher is better.
💡 A lightbulb marks a model its provider publishes as a reasoning model
The Known Good
Every model we hold a Mock AIME 2024–25 score for
All 238 of 238 models
| # | Model | Creator | Mock AIME 2024–25 | Known Good Index | Measured |
|---|---|---|---|---|---|
| 1 |
OpenAI: GPT-5.6 Sol
💡 |
OpenAI | 100.0 | — | 8 Oct 2026 |
| 2 |
Anthropic: Claude Fable 5
💡 |
Anthropic | 100.0 | — | 8 Oct 2026 |
| 3 |
gpt-5.5-pro-pre-release
|
OpenAI | 100.0 | — | 8 Oct 2026 |
| 4 |
gpt-5.5-pre-release
|
OpenAI | 100.0 | 97 | 8 Oct 2026 |
| 5 |
Anthropic: Claude Fable 5.1
💡 |
Anthropic | 100.0 | — | 8 Oct 2026 |
| 6 |
OpenAI: GPT-6 Astra
💡 |
OpenAI | 100.0 | — | 8 Oct 2026 |
| 7 |
Qwen: Qwen3.8 Max (0902)
💡 |
Alibaba | 100.0 | — | 8 Oct 2026 |
| 8 |
Anthropic: Claude Opus 5.5
💡 |
Anthropic | 100.0 | — | 8 Oct 2026 |
| 9 |
OpenAI: GPT-6 Sol
💡 |
OpenAI | 100.0 | — | 8 Oct 2026 |
| 10 |
Anthropic: Claude Sonnet 5.5
💡 |
Anthropic | 100.0 | — | 8 Oct 2026 |
| 11 |
OpenAI: GPT-6.1 Sol
💡 |
OpenAI | 100.0 | — | 8 Oct 2026 |
| 12 |
OpenAI: GPT-5.6 Terra
💡 |
OpenAI | 99.7 | — | 8 Oct 2026 |
| 13 |
Qwen: Qwen3.8 Max
💡 |
Alibaba | 99.4 | — | 8 Oct 2026 |
| 14 |
SpaceXAI: Grok 4.6
💡 |
xAI | 99.2 | — | 8 Oct 2026 |
| 15 |
Meta: Muse Spark 1.3
💡 |
Meta | 99.2 | — | 8 Oct 2026 |
| 16 |
Anthropic: Claude Opus 5
💡 |
Anthropic | 98.9 | — | 8 Oct 2026 |
| 17 |
Google: Gemini 3.8 Flash
💡 |
98.9 | — | 8 Oct 2026 | |
| 18 |
OpenAI: GPT-6 Luna
💡 |
OpenAI | 98.9 | — | 8 Oct 2026 |
| 19 |
DeepSeek: DeepSeek V4 Pro 0813
💡 |
DeepSeek | 98.6 | — | 8 Oct 2026 |
| 20 |
OpenAI: GPT-5.6 Luna
💡 |
OpenAI | 98.3 | — | 8 Oct 2026 |
| 21 |
Anthropic: Claude Opus 4.8
💡 |
Anthropic | 98.3 | — | 8 Oct 2026 |
| 22 |
SpaceXAI: Grok 4.7
💡 |
xAI | 98.1 | — | 8 Oct 2026 |
| 23 |
Anthropic: Claude Opus 4.7
💡 |
Anthropic | 97.8 | 90 | 8 Oct 2026 |
| 24 |
SpaceXAI: Grok 4.5
💡 |
xAI | 97.8 | — | 8 Oct 2026 |
| 25 |
OpenAI: GPT-5.4
💡 |
OpenAI | 97.8 | 89 | 8 Oct 2026 |
| 26 |
MoonshotAI: Kimi K3
💡 |
Moonshot AI | 97.2 | — | 8 Oct 2026 |
| 27 |
Google: Gemini 3.7 Flash
💡 |
97.2 | — | 8 Oct 2026 | |
| 28 |
DeepSeek: DeepSeek V4 Pro 0423
💡 |
DeepSeek | 96.7 | 93 | 8 Oct 2026 |
| 29 |
MoonshotAI: Kimi K2.6
💡 |
Moonshot AI | 96.1 | 92 | 8 Oct 2026 |
| 30 |
OpenAI: GPT-5.2
💡 |
OpenAI | 96.1 | 79 | 8 Oct 2026 |
| 31 |
Google: Gemini 3.1 Pro Preview
💡 |
95.6 | 93 | 8 Oct 2026 | |
| 32 |
MoonshotAI: Kimi K2.7 Code
💡 |
Moonshot AI | 95.6 | — | 8 Oct 2026 |
| 33 |
Qwen: Qwen3.7 Max
💡 |
Alibaba | 95.6 | 93 | 8 Oct 2026 |
| 34 |
Google: Gemini 3.5 Flash
💡 |
95.6 | 95 | 8 Oct 2026 | |
| 35 |
Google: Gemini 3 Flash Preview
💡 |
95.6 | 83 | 8 Oct 2026 | |
| 36 |
Anthropic: Claude Sonnet 5
💡 |
Anthropic | 94.7 | — | 8 Oct 2026 |
| 37 |
claude-opus-4-6_64K
|
Anthropic | 94.4 | — | 8 Oct 2026 |
| 38 |
DeepSeek: DeepSeek V4 Flash 0731
💡 |
DeepSeek | 94.4 | — | 8 Oct 2026 |
| 39 |
Google: Gemini 3.6 Flash
💡 |
94.2 | — | 8 Oct 2026 | |
| 40 |
Z.ai: GLM 5.3 Flash
💡 |
Z.ai | 93.9 | — | 8 Oct 2026 |
| 41 |
Qwen: Qwen3.7 Plus
💡 |
Alibaba | 93.3 | — | 8 Oct 2026 |
| 42 |
SpaceXAI: Grok 4.3
💡 |
xAI | 93.3 | — | 8 Oct 2026 |
| 43 |
Z.ai: GLM 5.1
💡 |
Z.ai | 93.3 | 90 | 8 Oct 2026 |
| 44 |
Qwen: Qwen3.6 Plus
💡 |
Alibaba | 93.3 | 79 | 8 Oct 2026 |
| 45 |
Anthropic: Claude Opus 4.6
💡 |
Anthropic | 93.1 | 87 | 8 Oct 2026 |
| 46 |
grok-4.20-0309-reasoning
|
xAI | 92.2 | — | 8 Oct 2026 |
| 47 |
fireworks/kimi-k2p5
|
Moonshot AI | 92.2 | — | 8 Oct 2026 |
| 48 |
OpenAI: GPT-5
💡 |
OpenAI | 91.4 | 77 | 8 Oct 2026 |
| 49 |
gemini-3-pro-preview
|
91.4 | 81 | 8 Oct 2026 | |
| 50 |
Qwen: Qwen3.6 Max Preview
💡 |
Alibaba | 91.1 | 89 | 8 Oct 2026 |
| 51 |
Qwen: Qwen3.6 27B
💡 |
Alibaba | 91.1 | — | 8 Oct 2026 |
| 52 |
Z.ai: GLM 5.3
💡 |
Z.ai | 91.1 | — | 8 Oct 2026 |
| 53 | TM Thinking Machines: Inkling Small 💡 | Thinking Machines | 90.0 | — | 8 Oct 2026 |
| 54 |
muse-spark
|
Meta | 88.9 | — | 8 Oct 2026 |
| 55 | TM Thinking Machines: Inkling 💡 | Thinking Machines | 88.9 | — | 8 Oct 2026 |
| 56 |
OpenAI: GPT-5.4 Mini
💡 |
OpenAI | 88.9 | — | 8 Oct 2026 |
| 57 |
Qwen: Qwen3.5 397B A17B
💡 |
Alibaba | 88.9 | — | 8 Oct 2026 |
| 58 |
OpenAI: gpt-oss-120b
💡 |
OpenAI | 88.9 | 62 | 8 Oct 2026 |
| 59 |
OpenAI: GPT-5.1
💡 |
OpenAI | 88.6 | 73 | 8 Oct 2026 |
| 60 |
deepseek-reasoner
|
DeepSeek | 87.8 | — | 8 Oct 2026 |
| 61 |
OpenAI: GPT-5.4 Nano
💡 |
OpenAI | 87.8 | — | 8 Oct 2026 |
| 62 |
NVIDIA: Nemotron 3 Ultra
💡 |
NVIDIA | 86.7 | — | 8 Oct 2026 |
| 63 |
Qwen: Qwen3.6 35B A3B
💡 |
Alibaba | 86.7 | 66 | 8 Oct 2026 |
| 64 |
OpenAI: GPT-5 Mini
💡 |
OpenAI | 86.7 | 66 | 8 Oct 2026 |
| 65 |
Qwen: Qwen3 235B A22B Thinking 2507
💡 |
Alibaba | 86.7 | — | 8 Oct 2026 |
| 66 |
qwen3.5-plus
|
Alibaba | 86.7 | — | 8 Oct 2026 |
| 67 |
Qwen: Qwen3.7 Flash
💡 |
Alibaba | 86.7 | — | 8 Oct 2026 |
| 68 |
Z.ai: GLM 5.2
💡 |
Z.ai | 86.4 | 91 | 8 Oct 2026 |
| 69 |
Anthropic: Claude Opus 4.5
💡 |
Anthropic | 86.1 | 71 | 8 Oct 2026 |
| 70 |
Anthropic: Claude Sonnet 4.6
💡 |
Anthropic | 85.8 | 80 | 8 Oct 2026 |
| 71 |
Qwen: Qwen3.6 Flash
💡 |
Alibaba | 84.4 | — | 8 Oct 2026 |
| 72 |
OpenAI: GPT-5.5
💡 |
OpenAI | 84.4 | 93 | 8 Oct 2026 |
| 73 |
Qwen: Qwen3.5-Flash
💡 |
Alibaba | 84.4 | — | 8 Oct 2026 |
| 74 |
OpenAI: o3
💡 |
OpenAI | 84.4 | 72 | 8 Oct 2026 |
| 75 |
claude-opus-4-8_none
|
Anthropic | 84.4 | — | 8 Oct 2026 |
| 76 |
Google: Gemini 2.5 Pro
💡 |
84.2 | 65 | 8 Oct 2026 | |
| 77 |
grok-4-0709
|
xAI | 84.0 | 68 | 8 Oct 2026 |
| 78 |
Z.ai: GLM 4.7
💡 |
Z.ai | 83.3 | 69 | 8 Oct 2026 |
| 79 |
kimi-k2-thinking-turbo
|
Moonshot AI | 83.1 | — | 8 Oct 2026 |
| 80 |
qwen3.5-397b-a17b_none
|
Alibaba | 82.2 | — | 8 Oct 2026 |
| 81 |
gemma-4-26b-a4b-it_minimal
|
82.2 | — | 8 Oct 2026 | |
| 82 |
OpenAI: o4 Mini
💡 |
OpenAI | 81.7 | — | 8 Oct 2026 |
| 83 |
claude-opus-4-5-20251101_16K
|
Anthropic | 81.7 | — | 8 Oct 2026 |
| 84 |
OpenAI: GPT-5 Nano
💡 |
OpenAI | 81.1 | 68 | 8 Oct 2026 |
| 85 |
Google: Gemini 3.1 Flash Lite
💡 |
80.0 | — | 8 Oct 2026 | |
| 86 |
Z.ai: GLM 5
💡 |
Z.ai | 80.0 | 77 | 8 Oct 2026 |
| 87 |
gemini-3.5-flash_minimal
|
80.0 | — | 8 Oct 2026 | |
| 88 |
gemini-3.6-flash_minimal
|
80.0 | — | 8 Oct 2026 | |
| 89 |
qwen3.7-plus_none
|
Alibaba | 80.0 | — | 8 Oct 2026 |
| 90 |
claude-sonnet-4-5-20250929_59K
|
Anthropic | 77.8 | — | 8 Oct 2026 |
| 91 |
claude-sonnet-4-5-20250929
|
Anthropic | 77.8 | 66 | 8 Oct 2026 |
| 92 |
grok-3-mini-beta
|
xAI | 77.8 | — | 8 Oct 2026 |
| 93 |
qwen3.7-flash_none
|
Alibaba | 77.8 | — | 8 Oct 2026 |
| 94 |
OpenAI: o3 Mini
💡 |
OpenAI | 76.9 | 86 | 8 Oct 2026 |
| 95 |
Qwen: Qwen3 Max
💡 |
Alibaba | 73.3 | — | 8 Oct 2026 |
| 96 |
OpenAI: o1
💡 |
OpenAI | 73.3 | 69 | 8 Oct 2026 |
| 97 |
gemma-4-31b-it_minimal
|
73.3 | — | 8 Oct 2026 | |
| 98 |
gemini-2.5-flash-preview-04-17
|
73.1 | — | 8 Oct 2026 | |
| 99 |
Google: Gemini 3.5 Flash Lite
💡 |
71.1 | — | 8 Oct 2026 | |
| 100 |
MiniMax: MiniMax M3
💡 |
MiniMax | 71.1 | — | 8 Oct 2026 |
| 101 |
claude-sonnet-4-5-20250929_16K
|
Anthropic | 71.1 | — | 8 Oct 2026 |
| 102 |
claude-sonnet-4-20250514
|
Anthropic | 71.1 | — | 8 Oct 2026 |
| 103 |
gemini-2.5-flash-preview-05-20
|
70.8 | — | 8 Oct 2026 | |
| 104 |
Qwen: Qwen3 30B A3B Thinking 2507
💡 |
Alibaba | 70.3 | — | 8 Oct 2026 |
| 105 |
qwen3.5-35b-a3b_none
|
Alibaba | 70.0 | — | 8 Oct 2026 |
| 106 |
claude-opus-4-1-20250805_27K
|
Anthropic | 68.9 | — | 8 Oct 2026 |
| 107 |
claude-sonnet-4-20250514_59K
|
Anthropic | 68.9 | — | 8 Oct 2026 |
| 108 |
gpt-5.6-sol_none
|
OpenAI | 68.9 | — | 8 Oct 2026 |
| 109 |
qwen3.6-35b-a3b_none
|
Alibaba | 68.9 | — | 8 Oct 2026 |
| 110 |
gpt-5.5-instant
|
OpenAI | 68.1 | — | 8 Oct 2026 |
| 111 |
seed-oss-36b-instruct
|
Unknown | 67.5 | — | 8 Oct 2026 |
| 112 |
Qwen: Qwen3 32B
💡 |
Alibaba | 66.9 | — | 8 Oct 2026 |
| 113 |
claude-haiku-4-5-20251001
|
Anthropic | 66.7 | 67 | 8 Oct 2026 |
| 114 |
qwen3.6-27b_none
|
Alibaba | 66.7 | — | 8 Oct 2026 |
| 115 |
DeepSeek: R1 0528
💡 |
DeepSeek | 66.4 | — | 8 Oct 2026 |
| 116 |
Qwen: Qwen3 14B
💡 |
Alibaba | 66.4 | — | 8 Oct 2026 |
| 117 |
OpenAI: gpt-oss-20b
💡 |
OpenAI | 65.3 | 42 | 8 Oct 2026 |
| 118 |
claude-opus-4-1-20250805_16K
|
Anthropic | 64.4 | — | 8 Oct 2026 |
| 119 |
claude-opus-4-20250514_27K
|
Anthropic | 64.4 | — | 8 Oct 2026 |
| 120 |
Qwen: Qwen3 30B A3B
💡 |
Alibaba | 62.8 | — | 8 Oct 2026 |
| 121 |
Qwen: Qwen3 30B A3B Instruct 2507
|
Alibaba | 62.2 | — | 8 Oct 2026 |
| 122 |
gpt-5.2-2025-12-11_none
|
OpenAI | 62.2 | — | 8 Oct 2026 |
| 123 |
qwen3.5-9B_none
|
Alibaba | 61.7 | — | 8 Oct 2026 |
| 124 |
claude-opus-4-20250514_16K
|
Anthropic | 60.0 | — | 8 Oct 2026 |
| 125 |
QwQ-32B
|
Alibaba | 59.2 | 69 | 8 Oct 2026 |
| 126 |
Z.ai: GLM 4.7 Flash
💡 |
Z.ai | 58.3 | — | 8 Oct 2026 |
| 127 |
claude-3-7-sonnet-20250219_64K
|
Anthropic | 57.8 | — | 8 Oct 2026 |
| 128 | GD gemini-2.0-flash-thinking-exp-01-21 | Google DeepMind,Google | 57.8 | 62 | 8 Oct 2026 |
| 129 |
gpt-5.4-2026-03-05_none
|
OpenAI | 57.8 | — | 8 Oct 2026 |
| 130 |
gpt-5.5_none
|
OpenAI | 57.8 | — | 8 Oct 2026 |
| 131 |
Qwen: Qwen3 8B
💡 |
Alibaba | 56.1 | — | 8 Oct 2026 |
| 132 |
qwen3.5-4b_none
|
Alibaba | 55.8 | — | 8 Oct 2026 |
| 133 |
grok-3-beta
|
xAI | 55.6 | — | 8 Oct 2026 |
| 134 |
DeepSeek-R1-Distill-Qwen-32B
|
DeepSeek | 55.6 | 53 | 8 Oct 2026 |
| 135 |
gpt-5-mini-2025-08-07_minimal
|
OpenAI | 55.6 | — | 8 Oct 2026 |
| 136 |
Qwen: Qwen3.5-35B-A3B
💡 |
Alibaba | 54.4 | — | 8 Oct 2026 |
| 137 |
DeepSeek: R1
💡 |
DeepSeek | 53.3 | 75 | 8 Oct 2026 |
| 138 |
claude-sonnet-4-20250514_16K
|
Anthropic | 53.3 | — | 8 Oct 2026 |
| 139 |
claude-3-7-sonnet-20250219
|
Anthropic | 53.3 | 74 | 8 Oct 2026 |
| 140 |
gpt-5.6-terra_none
|
OpenAI | 53.3 | — | 8 Oct 2026 |
| 141 |
qwen3-4b-instruct-2507
|
Unknown | 52.2 | — | 8 Oct 2026 |
| 142 |
DeepSeek: R1 Distill Llama 70B
💡 |
DeepSeek | 51.4 | 63 | 8 Oct 2026 |
| 143 |
gemini-3.5-flash-lite_minimal
|
51.1 | — | 8 Oct 2026 | |
| 144 |
DeepSeek-R1-Distill-Qwen-14B
|
DeepSeek | 50.6 | — | 8 Oct 2026 |
| 145 |
DeepSeek: DeepSeek V3
|
DeepSeek | 48.9 | 62 | 8 Oct 2026 |
| 146 |
o1-mini
|
OpenAI | 46.9 | 65 | 8 Oct 2026 |
| 147 |
claude-3-7-sonnet-20250219_16K
|
Anthropic | 46.7 | — | 8 Oct 2026 |
| 148 |
gpt-5-2025-08-07_minimal
|
OpenAI | 46.7 | — | 8 Oct 2026 |
| 149 |
gpt-5.4-nano-2026-03-17_none
|
OpenAI | 46.7 | — | 8 Oct 2026 |
| 150 |
deepseek-v4-pro_none
|
DeepSeek | 46.7 | — | 8 Oct 2026 |
| 151 |
OpenAI: GPT-4.1 Mini
|
OpenAI | 44.7 | — | 8 Oct 2026 |
| 152 |
deepseek-r1-0528-qwen3-8b
|
Unknown | 43.9 | — | 8 Oct 2026 |
| 153 |
Qwen: Qwen3.5-9B
💡 |
Alibaba | 42.8 | 44 | 8 Oct 2026 |
| 154 |
claude-opus-4-20250514
|
Anthropic | 42.2 | 68 | 8 Oct 2026 |
| 155 |
Anthropic: Claude Opus 4.1
💡 |
Anthropic | 40.0 | 49 | 8 Oct 2026 |
| 156 |
gpt-5.6-luna_none
|
OpenAI | 40.0 | — | 8 Oct 2026 |
| 157 |
OpenAI: GPT-4.1
|
OpenAI | 38.3 | 46 | 8 Oct 2026 |
| 158 |
DeepSeek: DeepSeek V3 0324
|
DeepSeek | 37.8 | 64 | 8 Oct 2026 |
| 159 |
gpt-4.5-preview
|
OpenAI | 37.8 | 54 | 8 Oct 2026 |
| 160 |
gpt-5.1-2025-11-13_none
|
OpenAI | 37.8 | — | 8 Oct 2026 |
| 161 |
gemini-3.1-flash-lite_minimal
|
37.8 | — | 8 Oct 2026 | |
| 162 |
gpt-5-nano-2025-08-07_minimal
|
OpenAI | 35.6 | — | 8 Oct 2026 |
| 163 |
Mistral: Mistral Medium 3
|
Mistral | 32.2 | — | 8 Oct 2026 |
| 164 | GD gemini-2.0-flash-001 | Google DeepMind,Google | 31.1 | 61 | 8 Oct 2026 |
| 165 |
o1-preview
|
OpenAI | 31.1 | — | 8 Oct 2026 |
| 166 |
mistral-small-3.2-2506
|
Mistral | 30.3 | — | 8 Oct 2026 |
| 167 |
magistral-small-2506
|
Mistral | 30.0 | — | 8 Oct 2026 |
| 168 |
OpenAI: GPT-4.1 Nano
|
OpenAI | 28.9 | — | 8 Oct 2026 |
| 169 |
glm-5.2_none
|
Z.ai | 28.9 | — | 8 Oct 2026 |
| 170 |
magistral-small-2509
|
Mistral | 28.1 | — | 8 Oct 2026 |
| 171 |
gpt-5.4-mini-2026-03-17_none
|
OpenAI | 26.7 | — | 8 Oct 2026 |
| 172 |
MiniMax-M3_none
|
MiniMax | 26.7 | — | 8 Oct 2026 |
| 173 |
qwen3-14b_none
|
Alibaba | 25.8 | — | 8 Oct 2026 |
| 174 |
Qwen3-30B-A3B_none
|
Alibaba | 25.6 | — | 8 Oct 2026 |
| 175 |
glm-4.7-flash_none
|
Unknown | 25.0 | — | 8 Oct 2026 |
| 176 |
gemini-1.5-pro-002
|
23.1 | — | 8 Oct 2026 | |
| 177 |
Qwen3-32B_none
|
Alibaba | 23.1 | — | 8 Oct 2026 |
| 178 |
Google: Gemma 3 27B
|
22.5 | 47 | 8 Oct 2026 | |
| 179 |
qwen3-8b_none
|
Alibaba | 22.2 | — | 8 Oct 2026 |
| 180 |
DeepSeek-R1-Distill-Qwen-1.5B
|
DeepSeek | 21.4 | — | 8 Oct 2026 |
| 181 |
Llama-4-Maverick-17B-128E-Instruct-FP8
|
Meta | 20.6 | — | 8 Oct 2026 |
| 182 |
Qwen: Qwen-Plus
💡 |
Alibaba | 17.8 | — | 8 Oct 2026 |
| 183 |
Google: Gemma 3 12B
|
16.7 | — | 8 Oct 2026 | |
| 184 |
gemini-1.5-flash-002
|
16.2 | — | 8 Oct 2026 | |
| 185 |
qwen-max
|
Alibaba | 16.1 | — | 8 Oct 2026 |
| 186 |
Microsoft: Phi 4
|
Microsoft | 13.8 | 41 | 8 Oct 2026 |
| 187 |
grok-2-1212
|
xAI | 11.5 | 45 | 8 Oct 2026 |
| 188 |
Llama-3.1-405B-Instruct
|
Meta | 9.7 | — | 8 Oct 2026 |
| 189 |
Mistral Large 2407
|
Mistral | 8.5 | — | 8 Oct 2026 |
| 190 |
claude-3-5-sonnet-20241022
|
Anthropic | 8.5 | 45 | 8 Oct 2026 |
| 191 |
Qwen2.5 72B Instruct
|
Alibaba | 8.1 | — | 8 Oct 2026 |
| 192 |
qwen3-1.7b_none
|
Alibaba | 8.1 | — | 8 Oct 2026 |
| 193 |
Llama-4-Scout-17B-16E-Instruct
|
Meta | 7.8 | — | 8 Oct 2026 |
| 194 |
mistral-large-2411
|
Mistral | 7.8 | 37 | 8 Oct 2026 |
| 195 |
Google: Gemma 3 4B
|
7.5 | — | 8 Oct 2026 | |
| 196 |
qwen2.5-32b-instruct
|
Alibaba | 7.4 | — | 8 Oct 2026 |
| 197 |
OpenAI: GPT-4o-mini
|
OpenAI | 6.9 | 31 | 8 Oct 2026 |
| 198 |
gemini-1.5-pro-001
|
6.8 | — | 8 Oct 2026 | |
| 199 |
OpenAI: GPT-4 Turbo
|
OpenAI | 6.7 | — | 8 Oct 2026 |
| 200 |
mistral-small-3-2501
|
Mistral | 6.7 | — | 8 Oct 2026 |
| 201 |
claude-3-5-sonnet-20240620
|
Anthropic | 6.5 | — | 8 Oct 2026 |
| 202 |
OpenAI: GPT-4o
|
OpenAI | 6.4 | 27 | 8 Oct 2026 |
| 203 |
qwen-turbo
|
Alibaba | 6.1 | — | 8 Oct 2026 |
| 204 |
mistral-small-2503
|
Mistral | 5.8 | 33 | 8 Oct 2026 |
| 205 |
mistral-small-2501
|
Mistral | 5.3 | 31 | 8 Oct 2026 |
| 206 |
Meta: Llama 3.3 70B Instruct
|
Meta | 5.1 | 34 | 8 Oct 2026 |
| 207 |
claude-3-opus-20240229
|
Anthropic | 4.7 | 32 | 8 Oct 2026 |
| 208 |
gemini-1.5-flash-8b-001
|
4.6 | — | 8 Oct 2026 | |
| 209 | AI Llama-3.1-Tulu-3-70B-DPO | Allen Institute for AI,University of Washington | 4.4 | — | 8 Oct 2026 |
| 210 |
claude-3-5-haiku-20241022
|
Anthropic | 4.3 | 30 | 8 Oct 2026 |
| 211 |
Meta-Llama-3-70B-Instruct
|
Meta | 4.3 | — | 8 Oct 2026 |
| 212 |
gemini-1.5-flash-001
|
3.9 | — | 8 Oct 2026 | |
| 213 |
mistral-small-3.1-2503
|
Mistral | 3.9 | — | 8 Oct 2026 |
| 214 |
Meta: Llama 3.1 70B Instruct
|
Meta | 3.6 | — | 8 Oct 2026 |
| 215 |
IBM: Granite 4.0 Micro
|
IBM | 2.8 | — | 8 Oct 2026 |
| 216 |
Llama-3.2-90B-Vision-Instruct
|
Meta | 2.6 | — | 8 Oct 2026 |
| 217 |
Qwen: Qwen2.5 7B Instruct
|
Alibaba | 2.5 | — | 8 Oct 2026 |
| 218 |
claude-2.0
|
Anthropic | 2.5 | — | 8 Oct 2026 |
| 219 |
claude-3-sonnet-20240229
|
Anthropic | 2.5 | — | 8 Oct 2026 |
| 220 | NR Hermes-2-Theta-Llama-3-70B | Nous Research,Arcee AI | 2.5 | — | 8 Oct 2026 |
| 221 |
gpt-3.5-turbo-0125
|
OpenAI | 2.2 | — | 8 Oct 2026 |
| 222 |
claude-2.1
|
Anthropic | 1.9 | — | 8 Oct 2026 |
| 223 |
Meta-Llama-3-8B-Instruct
|
Meta | 1.9 | — | 8 Oct 2026 |
| 224 |
mistral-large-2402
|
Mistral | 1.9 | — | 8 Oct 2026 |
| 225 |
Anthropic: Claude 3 Haiku
|
Anthropic | 1.8 | — | 8 Oct 2026 |
| 226 |
Meta: Llama 3.1 8B Instruct
💡 |
Meta | 1.7 | — | 8 Oct 2026 |
| 227 |
Google: Gemma 2 27B
|
1.4 | 21 | 8 Oct 2026 | |
| 228 |
gpt-4-0613
|
OpenAI | 1.1 | — | 8 Oct 2026 |
| 229 |
gemini-1.0-pro-001
|
1.1 | — | 8 Oct 2026 | |
| 230 |
gemma-3-1b-it
|
1.1 | — | 8 Oct 2026 | |
| 231 |
deepseek-llm-67b-chat
|
DeepSeek | 0.8 | — | 8 Oct 2026 |
| 232 |
granite-4.0-1b
|
Unknown | 0.8 | — | 8 Oct 2026 |
| 233 |
Meta: Llama 3.2 1B Instruct
|
Meta | 0.6 | — | 8 Oct 2026 |
| 234 |
gpt-4-0314
|
OpenAI | 0.6 | — | 8 Oct 2026 |
| 235 |
gemma-2-9b-it
|
0.6 | 13 | 8 Oct 2026 | |
| 236 |
Mistral-7B-Instruct-v0.3
|
Mistral | 0.3 | — | 8 Oct 2026 |
| 237 |
granite-4.0-350m
|
Unknown | 0.3 | — | 8 Oct 2026 |
| 238 |
Llama-2-70b-chat-hf
|
Meta | 0.0 | — | 8 Oct 2026 |
The Known Good
Provenance. These figures are published by Epoch AI and redistributed here under CC-BY 4.0.
The publisher's own page is here. We did not run this evaluation and we do not adjust the published numbers — where a score feeds an index it is
min-max normalised against every other model holding the same evaluation, and nothing else is done to it.
This evaluation is one of the seven components of the
Known Good Index. Full licence detail is on attribution, and every score here is
in the CSV download.