TAU3-BenchBrowse 284 families
TAU3-Bench
Can the model complete substantial programming work under this benchmark's agent setup?
Version / measure
- Domain
- Coding
- Ranked models
- 5
- Reported values
- 5
- Reports
- 1 · lab
- Direction
- Higher is better
- Status
- Version not specified
Rankings
Best reported score per model across effort settings
| # | Model | Best setting | Score |
|---|---|---|---|
| 1 | Qwen3.5-35B-A3BQwen | reportedcost n/a1 value · Apr 15 · source | 68.9% |
| 2 | Qwen3.5-27BQwen | reportedcost n/a1 value · Apr 15 · source | 68.4% |
| 3 | Gemma4-31BGoogle | reportedcost n/a1 value · Apr 15 · source | 67.5% |
| 4 | Qwen3.6-35B-A3BQwen | reportedcost n/a1 value · Apr 15 · source | 67.2% |
| 5 | Gemma4-26B-A4BGoogle | reportedcost n/a1 value · Apr 15 · source | 59.0% |
Effort curve
Score vs. reported cost. Lines connect complete sweeps only.
No cost-linked effort sweep for this version.
Definition and comparison boundarypublic methodology
Can the model complete substantial programming work under this benchmark's agent setup?
Software implementation, debugging, or repository work under the published evaluation protocol. Higher is better. The value is the percentage reported in this lab's table.
Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.
Official user model GPT-5.2 at low reasoning effort plus default BM25 retrieval.
Method / source ↗