Laboratory Science Knowledge Evaluation on LAB-Bench 2 821 (Evaluation)
82.3AccuracyClaude Opus 4.6
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude Opus 4.6Type=Commercial LLMs, Setting=BioMedArena2026.05 | 82.3 | |
| SOTASource=[11]2026.05 | 80 | |
| Claude Sonnet 4.6Type=Commercial LLMs, Setting=BioMedArena2026.05 | 74.3 | |
| Claude Opus 4.5Type=Commercial LLMs, Setting=BioMedArena2026.05 | 73.8 | |
| GPT-5.4Type=Commercial LLMs, Setting=BioMedArena2026.05 | 72.1 | |
| Claude Sonnet 4.5Type=Commercial LLMs, Setting=BioMedArena2026.05 | 71.5 | |
| Gemini 3.1 ProType=Commercial LLMs, Setting=BioMedArena2026.05 | 70.9 | |
| Nemotron-3 Super 120BType=Public LLMs, Setting=BioMedArena2026.05 | 62.6 | |
| GLM-4.5Type=Public LLMs, Setting=BioMedArena2026.05 | 61.5 | |
| Trinity-LargeType=Public LLMs, Setting=BioMedArena2026.05 | 61 | |
| Gemini 3 FlashType=Commercial LLMs, Setting=BioMedArena2026.05 | 59.8 | |
| INTELLECT-3.1Type=Public LLMs, Setting=BioMedArena2026.05 | 57.9 | |
| Qwen3-235BType=Public LLMs, Setting=BioMedArena2026.05 | 57.9 | |
| Gemini 3.1 ProType=Commercial LLMs, Setting=Baseline2026.05 | 52.1 | |
| Gemini 3 FlashType=Commercial LLMs, Setting=Baseline2026.05 | 51.9 | |
| Claude Opus 4.6Type=Commercial LLMs, Setting=Baseline2026.05 | 51.2 | |
| Claude Opus 4.5Type=Commercial LLMs, Setting=Baseline2026.05 | 49.8 | |
| Claude Sonnet 4.6Type=Commercial LLMs, Setting=Baseline2026.05 | 49.3 | |
| GPT-5.4Type=Commercial LLMs, Setting=Baseline2026.05 | 48.5 | |
| Claude Sonnet 4.5Type=Commercial LLMs, Setting=Baseline2026.05 | 48.1 | |
| GLM-4.5Type=Public LLMs, Setting=Baseline2026.05 | 44.8 | |
| Nemotron-3 Super 120BType=Public LLMs, Setting=Baseline2026.05 | 43.6 | |
| Trinity-LargeType=Public LLMs, Setting=Baseline2026.05 | 43.5 | |
| INTELLECT-3.1Type=Public LLMs, Setting=Baseline2026.05 | 43.5 | |
| Qwen3-235BType=Public LLMs, Setting=Baseline2026.05 | 42.5 |