Retail · Standard · pass^1
τ-bench · 8 models
| Rank | Model | Score |
|---|---|---|
| 1 | Qwen3.5-397B-A17B | 84.43 |
| 2 | GPT-5.2 | 81.58 |
| 3 | Claude Opus 4.5 | 79.61 |
| 4 | Gemini 3 Flash | 76.75 |
| 5 | Gemini 3 Pro | 75.88 |
| 6 | GPT-5.2 | 75 |
| 7 | GLM-5 | 73.68 |
| 8 | Claude Sonnet 4.5 | 72.37 |
τ-bench · 8 models
| Rank | Model | Score |
|---|---|---|
| 1 | Qwen3.5-397B-A17B | 84.43 |
| 2 | GPT-5.2 | 81.58 |
| 3 | Claude Opus 4.5 | 79.61 |
| 4 | Gemini 3 Flash | 76.75 |
| 5 | Gemini 3 Pro | 75.88 |
| 6 | GPT-5.2 | 75 |
| 7 | GLM-5 | 73.68 |
| 8 | Claude Sonnet 4.5 | 72.37 |