Wikibench

ZClawBench

coding

Multilingual: no

ZClawBench evaluates Claw-style agent task execution quality, measuring a model's ability to autonomously complete complex multi-step coding tasks in real-world environments.

Methodology

Imported from llm-stats public benchmark metadata. Modality: text. Max score: 1. Categories: agents, coding. Language: en. Verified by llm-stats: no.

Leaderboard

Showing 3 of 3 results

  1. Qwen3.7 Max self-reported llm-stats
    64.3%
  2. GLM-5V-Turbo self-reported llm-stats
    57.6%
  3. Qwen3.6-27B self-reported llm-stats
    53.4%