Open LLM Leaderboard - BBH: leaderboard

The "BBH" subtask of Open LLM Leaderboard. A task from the Open LLM Leaderboard, the standard open-weight evaluation suite spanning knowledge, reasoning, math and truthfulness benchmarks.

Metric: Score. Source: huggingface.co. Saturation forecast: Estimated already saturated. 4576 models tracked.

Top models

#ModelScoreOverall rank
1internlm2.5-20B-chat74.74#788
2calme-3.2-instruct-78B73.19#354
3Qwen 2.5 72B Instruct72.73#470
4InternLM2.5-Chat-7B70.73#1010
5Rombos-LLM-V2.5-Qwen-32B70.46#371
6oxyge1-33B70.33#697
7Qwen 2 72B Instruct69.77#580
8Awqward2.5-32B-Instruct69.74#370
9Qwentile2.5-32B-Instruct69.63#345
10Qwen 2 VL 72B Instruct69.46#551
11Qwen2.5-32B-Instruct-abliterated-v269.34#400
12Llama 3.3 70B Instruct69.19#541
13Llama 3.1 70B Instruct69.17#574
14Qwen 2.5 32B Instruct69.13#511
15solar-pro-preview-instruct68.17#755

No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #6 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=open-llm-leaderboard-bbh · How It Works · Data refreshed daily, snapshot 2026-11-06.