Binding Affinity Comparison on MolBench-MS
81.1AccuracyMolClaw (Claude Code)
Evaluation Results
| Method | Links | |
|---|---|---|
| MolClaw (Claude Code)Model Category=Agent, LLM Backend=Claude Sonnet 4.62026.04 | 81.1 | |
| MolClaw (OpenClaw)Model Category=Agent, LLM Backend=Claude Sonnet 4.62026.04 | 73 | |
| Gemini 3Model Category=LLM2026.04 | 56.8 | |
| Claude Sonnet 4.6Model Category=LLM2026.04 | 56.8 | |
| Kimi 2.5Model Category=LLM2026.04 | 51.4 | |
| Qwen 3.5Model Category=LLM2026.04 | 51.4 | |
| Claude Code (vanilla)Model Category=Agent, LLM Backend=Claude Sonnet 4.62026.04 | 51.4 | |
| OpenClaw (vanilla)Model Category=Agent, LLM Backend=Claude Sonnet 4.62026.04 | 51.4 | |
| GPT 5.2Model Category=LLM2026.04 | 48.7 | |
| Deepseek V3.2Model Category=LLM2026.04 | 43.2 | |
| GLM 5Model Category=LLM2026.04 | 24.3 | |
| BiomniModel Category=Agent2026.04 | 24.3 | |
| Minimax 2.5Model Category=LLM2026.04 | 16.2 |