Benchmark
WildClawBench
What is WildClawBench?
WildClawBench is an agentic coding benchmark from InternLM/Claw-Eval that reports overall model performance on real-world tool-using development tasks.
- Modality
- text
- Evaluates
- agents, coding
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 5
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/wildclawbench?top_n=500
WildClawBench results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Hy3 | Tencent | 0.536 | 2026-07-06 | evidence |
| MiMo-V2.5-Pro | Xiaomi | 0.43 | 2026-04-27 | evidence |
| Muse Glimmer-30B | Meta | 0.476 | 2026-08-10 | evidence |
| Seed 2.1 Pro | ByteDance | 0.617 | 2026-06-24 | evidence |
| Seed 2.1 Turbo | ByteDance | 0.628 | 2026-06-24 | 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.