τ2-Bench

Explore score history ↗

τ2-Bench is an AI benchmark in the Benchmark Radar catalog.

Released: 2025-06-10
Openness: unknown
Importer: Benchmark Saturation

Source provenance

Original evidence

τ2-Bench results and reported scores

Reported benchmark scores · Benchmark Saturation

Reported scores: 5

Model	Organization	Reported value	Reported	Evidence
Claude-Sonnet-4.5	—	84.7	—	evidence
DeepSeek-V3.2	—	80.4	—	evidence
Gemini-3.0-Pro	—	85.4	—	evidence
gpt 5	—	80.0	—	evidence
Qwen3-Max-Thinking-Preview	—	80.8	—	evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.

Which sources cite τ2-Bench?

τ2-Bench