Benchmark
MMLU-Pro
Scores are percentages. Higher is better.
- Models and systems measured
- 299
- Scores recorded
- 328
- Sources
- 5
- IQ weight
- 1.01
Score distribution
- 54.6%
- Mean
- 55.3%
- Median
- 26.8%
- Std dev
- 17.2%
- 10th percentile
- 85.2%
- 90th percentile
- 0.649%
- Min
This benchmark has a weight of 1.01 for chat models, measured over 292 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model or system | Score |
|---|---|---|
| 151 | Llama 3.1 405B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim | 61.6% |
| 152 | Gemma 3 12B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 60.6% |
| 153 | Hunyuan Large· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim | 60.2% |
| 154 | Gemma 4 E2B-it· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 60% |
| 155 | Gemini 1.5 Flash-8B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 58.7% |
| 156 | arcee-blitz· Hugging Face Open LLM Leaderboard data | 57.3% |
| 157 | Claude 3 Sonnet· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 56.8% |
| 158 | Qwen2.5 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 56.3% |
| 159 | Granite 4.1 8B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim | 56% |
| 160 | Phi-3 Medium 4K Instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 55.7% |
| 161 | DeepSeek V2 Chat· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim | 54.8% |
| 162 | Jamba 1.5 Large· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 53.5% |
| 163 | EVA-Qwen2.5-72B-v0.2· Hugging Face Open LLM Leaderboard data | 53.5% |
| 164 | Phi 4 Mini· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 3 sources
| 52.8% |
| 165 | Meta Llama 3 70b· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 52.8% |
| 166 | Qwen 2 (72B)· Hugging Face Open LLM Leaderboard data | 52.6% |
| 167 | Llama 3.1 70B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 52.5% |
| 168 | Qwen2-VL (72B) Instruct· Hugging Face Open LLM Leaderboard data | 52.4% |
| 169 | Yi 1.5 34B Chat· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 52.3% |
| 170 | QwQ 32B Preview· Hugging Face Open LLM Leaderboard data | 52% |
| 171 | Phi-3 Medium 128K Instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 51.9% |
| 172 | Gemma 3n E4B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 50.6% |
| 173 | Qwen 2.5 Coder 32B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 50.4% |
| 174 | L3.3 Ms Nevoria 70b· Hugging Face Open LLM Leaderboard data | 50.4% |
| 175 | dolphin-2-9-2-qwen2-72b· Hugging Face Open LLM Leaderboard data | 49.7% |
| 176 | L3.3 Nevoria R1 70b· Hugging Face Open LLM Leaderboard data | 49.6% |
| 177 | Magnum V2 72B· Hugging Face Open LLM Leaderboard data | 49.5% |
| 178 | Qwen2 72B Dracarys· Hugging Face Open LLM Leaderboard data | 49.5% |
| 179 | Llama 3.1 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 3 sources
| 48.3% |
| 180 | GLM 4 9B 0414· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 47.9% |
| 181 | Qwen 2.5 Coder 32B· Hugging Face Open LLM Leaderboard data | 47.8% |
| 182 | Qwen2.5-Omni 7B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 47% |
| 183 | Llama 3 70B Instruct· Hugging Face Open LLM Leaderboard data | 46.7% |
| 184 | Phi-3 Mini 4K Instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 45.7% |
| 185 | l3-70b-euryale-v2.1· Hugging Face Open LLM Leaderboard data | 45.6% |
| 186 | Aya Expanse 32B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 45.4% |
| 187 | phi-3.5-moe-instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 45.3% |
| 188 | gemma-2-9b· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 45.1% |
| 189 | Qwen2.5 7B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 45% |
| 190 | Qwen2 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 44.1% |
| 191 | Phi-3 Mini 128K Instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 43.9% |
| 192 | Gemma 3 4B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 43.6% |
| 193 | Llama 3.1 Nemotron 70B Instruct HF· Hugging Face Open LLM Leaderboard data | 43.5% |
| 194 | Yi-34B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 43% |
| 195 | Jamba 1.5 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 42.5% |
| 196 | Qwen1.5-110B-Chat· Hugging Face Open LLM Leaderboard data | 42.5% |
| 197 | Qwen2.5 Math 72B Instruct· Hugging Face Open LLM Leaderboard data | 42.4% |
| 198 | Qwen3.5 0.8B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 42.3% |
| 199 | DeepSeek R1 Distill Llama 70B· Hugging Face Open LLM Leaderboard data | 41.6% |
| 200 | deepseek-coder-v2-lite-instruct· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim | 41.6% |
Showing 151–200 of 322 results.
Not ranked
The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.
| Model or system | Score |
|---|---|
| DeepSeek V4.1 Flash BaseEM · 5-shot· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 74.1% |
| MiMo V2 Flash Baseofficial Base table · 5-shot· Reported by Xiaomi MiMo· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 73.2% |
| Qwen3.5-397B-A17B-BaseQwen3.5 release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 76% |
| Qwen3.7-Plus-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 70.9% |
| Qwen3.8-27B-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 68.6% |
| Qwen3.8-Flash-Next-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 73.2% |
Data credits: Hugging Face official benchmark leaderboards; licence varies by benchmark · Data by LLM Stats · OpenEvals/leaderboard-data, MIT (licence) · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.