Skip to content

Benchmark

MMLU

Scores are percentages. Higher is better.

Models and systems measured
138
Scores recorded
153
Sources
3
IQ weight
0.00

Score distribution

76.9%
Mean
80.1%
Median
12.2%
Std dev
60.1%
10th percentile
88.8%
90th percentile
36.6%
Min
36.6%92.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.00 for chat models, measured over 133 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 systemScore
51qwen3-vl-30b· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.85%
52Phi 4· Epoch AI Benchmarking Hub data
Values from 2 sources
84.8%
53Llama 3.1 405B· Epoch AI Benchmarking Hub data84.4%
54GPT OSS 20bofficial model card Table 3 · reasoning medium · published setting; Table 3 does not further specify tool access · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.84%
55Mistral Large 2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim84%
56Llama 3.1 70B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
83.6%
57Llama 3.1 70B Instruct5-shot · macro_avg/acc· Reported by Meta· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.83.6%
58Qwen2.5 32B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim83.3%
59Qwen2 72B Instruct· Epoch AI Benchmarking Hub data
Values from 2 sources
82.4%
60GPT-4o mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
82%
61Nova Prov1· Epoch AI Benchmarking Hub data82%
62Jamba 1.5 Large· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim81.2%
63Qwen3 VL 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it.80.7%
64Mistral Small 3.1 24b Instruct 2503· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.6%
65Mistral Small 3.2 24B Instruct2506· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.5%
66Nova Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.5%
67DeepSeek V2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.4%
68GPT OSS 20bofficial model card Table 3 · reasoning low · published setting; Table 3 does not further specify tool access · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.80.4%
69Llama 3.2 90B Vision Instruct· Epoch AI Benchmarking Hub data80.3%
70Llama 3.1 Nemotron 70B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.2%
71GPT-4.1 nano· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim80.1%
72Mistral Large2407· Epoch AI Benchmarking Hub data80%
73Qwen 2.5 14B Instruct· Epoch AI Benchmarking Hub data
Values from 2 sources
79.9%
74Gemini 2.0 Flash Experimental· Epoch AI Benchmarking Hub data79.7%
75Llama 4 Scout· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79.6%
76Llama 3 70B Instruct· Epoch AI Benchmarking Hub data79.3%
77yi-large· Epoch AI Benchmarking Hub data79.3%
78Qwen 2.5 Coder 32B· Epoch AI Benchmarking Hub data79.1%
79Claude 3 Sonnet· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
79%
80Gemini 1.5 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
78.9%
81phi-3.5-moe-instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim78.9%
82Claude 2· Epoch AI Benchmarking Hub data78.5%
83Phi-3 Medium 128K Instruct· Epoch AI Benchmarking Hub data78%
84gemini-1.5-flash-001· Epoch AI Benchmarking Hub data77.9%
85mixtral-8x22b-v0.1· Epoch AI Benchmarking Hub data77.8%
86Nova Micro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim77.6%
87Nova Litev1· Epoch AI Benchmarking Hub data77%
88Yi-34B· Epoch AI Benchmarking Hub data76.3%
89Command R Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75.7%
90Gemma 2 27B IT· Epoch AI Benchmarking Hub data
Values from 2 sources
75.7%
91Phi-3 Small 8K Instruct· Epoch AI Benchmarking Hub data75.7%
92Claude 3 Haiku· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
75.2%
93qwen2p5-coder-14b· Epoch AI Benchmarking Hub data75.2%
94Qwen 2.5 Coder 32B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75.1%
95Claude Haiku 3.5· Epoch AI Benchmarking Hub data74.3%
96gemini-1.5-flash-002· Epoch AI Benchmarking Hub data73.9%
97Claude 2.1· Epoch AI Benchmarking Hub data73.5%
98Yi 34B Chat· Epoch AI Benchmarking Hub data73.5%
99Claude Instant 1.2· Epoch AI Benchmarking Hub data73.2%
100Llama 3.2 11B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73%

Showing 51–100 of 148 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 systemScore
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.86.7%
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.88.6%
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.90.4%
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.87.5%
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.90.4%

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · Data by LLM Stats · 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.