BBH
Big-Bench Hard benchmark with challenging tasks requiring multi-step reasoning.
16rows
scoreprimary metric
2026-05-06sampled
Metadata
Metrics
Score, Standard error (lower is better)
| Rank | Subject | Score | Model Match | Provenance | Sampled |
|---|---|---|---|---|---|
| 1 | DeepSeek V3 | 87.50 |  DeepSeek V3 deepseek-deepseek-chat | Imported | 2026-05-06 |
| 2 | Llama 3.1 405B | 82.90 | — | Imported | 2026-05-06 |
| 3 | Phi-3-medium-128k-instruct | 81.40 | — | Imported | 2026-05-06 |
| 4 | Qwen 2.5 72B | 79.80 |  Qwen2.5 72B Instruct qwen-qwen-2.5-72b-instruct | Imported | 2026-05-06 |
| 5 | Phi-3-small-8k-instruct | 79.10 | — | Imported | 2026-05-06 |
| 6 | GPT-4.1 | 75.12 |  GPT-4.1 openai-gpt-4.1 | Imported | 2026-05-06 |
| 7 | Phi-3-mini-4k-instruct | 71.70 | — | Imported | 2026-05-06 |
| 8 | Llama-2-70b-hf | 64.90 | — | Imported | 2026-05-06 |
| 9 | gpt-3.5-turbo-1106 | 61.59 | GPT-3.5 Turbo openai-gpt-3.5-turbo | Imported | 2026-05-06 |
| 10 | Phi-4 | 59.40 |  Phi 4 microsoft-phi-4 | Imported | 2026-05-06 |
| 11 | Llama-2-7b | 58.50 | — | Imported | 2026-05-06 |
| 12 | Mistral-7B-v0.1 | 56.10 | — | Imported | 2026-05-06 |
| 13 | gemma-7b | 55.10 | — | Imported | 2026-05-06 |
| 14 | Qwen 3 235B | 55 | Qwen3 235B A22B qwen-qwen3-235b-a22b | Imported | 2026-05-06 |
| 15 | Yi-6B | 47.20 | — | Imported | 2026-05-06 |
| 16 | falcon-180B | 37.10 | — | Imported | 2026-05-06 |
No matching rows.