Retail · Standard · pass^4
τ-bench · 8 models
| Rank | Model | Score |
|---|---|---|
| 1 | Qwen3.5-397B-A17B | 59.65 |
| 2 | Claude Opus 4.5 | 51.75 |
| 2 | GPT-5.2 | 51.75 |
| 2 | Gemini 3 Flash | 51.75 |
| 5 | Gemini 3 Pro | 47.37 |
| 6 | GPT-5.2 | 45.614 |
| 7 | GLM-5 | 43.86 |
| 8 | Claude Sonnet 4.5 | 39.47 |
τ-bench · 8 models
| Rank | Model | Score |
|---|---|---|
| 1 | Qwen3.5-397B-A17B | 59.65 |
| 2 | Claude Opus 4.5 | 51.75 |
| 2 | GPT-5.2 | 51.75 |
| 2 | Gemini 3 Flash | 51.75 |
| 5 | Gemini 3 Pro | 47.37 |
| 6 | GPT-5.2 | 45.614 |
| 7 | GLM-5 | 43.86 |
| 8 | Claude Sonnet 4.5 | 39.47 |