LLM Stats (TAU-bench Retail): leaderboard
Metric: Score (%). Source: llm-stats.com. 25 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.1 | 82.4 |
| 2 | Claude Opus 4 | 81.4 |
| 3 | Claude 3.7 Sonnet | 81.2 |
| 4 | Claude Sonnet 4 | 80.5 |
| 5 | GLM-4.5 | 79.7 |
| 6 | GLM 4.5 Air | 77.9 |
| 7 | Qwen 3 Coder 480B A35B Instruct | 77.5 |
| 8 | O4 Mini | 71.8 |
| 9 | O1 | 70.8 |
| 10 | Qwen 3 Next 80B A3B (Thinking) | 69.6 |
| 11 | Claude 3.5 Sonnet | 69.2 |
| 12 | GPT-4.5 | 68.4 |
| 13 | GPT-4.1 | 68 |
| 14 | GPT-OSS-120B | 67.8 |
| 15 | Qwen 3 235B A22B 2507 (Thinking) | 67.8 |
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-tau-bench-retail · How It Works · Data refreshed daily, snapshot 2026-09-19.