Benchmark
t2-bench
What is t2-bench?
t2-bench is a benchmark for evaluating agentic tool use capabilities, measuring how well models can select, sequence, and utilize tools to solve complex tasks…
- Modality
- text
- Evaluates
- reasoning, agents, tool calling
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 23
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/t2-bench?top_n=500
t2-bench results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| DeepSeek-V3.2 | DeepSeek | 0.803 | 2025-12-01 | evidence |
| DeepSeek-V3.2 (Thinking) | DeepSeek | 0.802 | 2025-12-01 | evidence |
| DeepSeek-V3.2-Speciale | DeepSeek | 0.803 | 2025-12-01 | evidence |
| DiffusionGemma 26B-A4B | 0.562 | 2026-06-10 | evidence | |
| Gemini 3 Flash | 0.902 | 2025-12-17 | evidence | |
| Gemini 3 Pro | 0.854 | 2025-11-18 | evidence | |
| Gemini 3.1 Pro | 0.993 | 2026-02-19 | evidence | |
| Gemma 4 26B-A4B | 0.855 | 2026-04-02 | evidence | |
| Gemma 4 31B | 0.864 | 2026-04-02 | evidence | |
| Gemma 4 E2B | 0.294 | 2026-04-02 | evidence | |
| Gemma 4 E4B | 0.575 | 2026-04-02 | evidence | |
| GLM-5 | Z.ai | 0.897 | 2026-02-11 | evidence |
| GPT OSS 120B High | OpenAI | 0.639 | 2025-08-05 | evidence |
| K-EXAONE-236B-A23B | LG AI Research | 0.732 | 2025-12-31 | evidence |
| Qwen3 Max | Qwen | 0.748 | 2025-12-15 | evidence |
| Qwen3.5-0.8B | Qwen | 0.116 | 2026-03-02 | evidence |
| Qwen3.5-122B-A10B | Qwen | 0.795 | 2026-02-24 | evidence |
| Qwen3.5-27B | Qwen | 0.79 | 2026-02-24 | evidence |
| Qwen3.5-2B | Qwen | 0.488 | 2026-03-02 | evidence |
| Qwen3.5-35B-A3B | Qwen | 0.812 | 2026-02-24 | evidence |
| Qwen3.5-397B-A17B | Qwen | 0.867 | 2026-02-16 | evidence |
| Qwen3.5-4B | Qwen | 0.799 | 2026-03-02 | evidence |
| Qwen3.5-9B | Qwen | 0.791 | 2026-03-02 | 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.