Benchmark leaderboard
WildClawBench OpenClaw is a public benchmark board published by WildClawBench. ModelCap records each tracked model's published score and position on it without re-running the evaluation. This page lists every model ModelCap tracks with a published WildClawBench OpenClaw result, its best evaluated configuration, the source's own rank, and the model's ModelCap Index position, API price and context window.
Snapshot as of 20 July 2026
GPT-5.6 Sol from OpenAI leads the WildClawBench OpenClaw leaderboard among the 24 tracked models with 67.2, per the source snapshot published 20 July 2026.
Sorted by published score, one row per model. Source: WildClawBench.
WildClawBench OpenClaw ranks models by the published WildClawBench OpenClaw score. ModelCap ingests the board as published, matches each entry to a catalogue model with a reviewed identity, and shows the source's score, interval and rank unchanged. Where a model has an admitted result, it feeds the ModelCap Index as agent evidence alongside the other public boards; the methodology documents the weighting and the identity rules.
WildClawBench OpenClaw is a public benchmark board published by WildClawBench. ModelCap records each tracked model's published score and position on it without re-running the evaluation. It is published by WildClawBench.
GPT-5.6 Sol (OpenAI) holds the top WildClawBench OpenClaw score among the models ModelCap tracks, at 67.2 as of the source snapshot published 20 July 2026.
24 tracked models have a published WildClawBench OpenClaw result on ModelCap; the source board itself lists 34 entries. Each row shows the model's best evaluated configuration.
GLM 5.2 from Z.ai is the highest-scoring model with openly downloadable weights on this board, at 54.2.
Among the ten highest-scoring priced models, GLM 5.2 has the lowest listed output price at $3.15 per 1M tokens while scoring 54.2.
The board ranks models by the published WildClawBench OpenClaw score; higher scores are better. ModelCap shows the source's own score, interval and rank and never re-runs the evaluation.
Not on its own. The ModelCap Index combines several public capability sources with published uncertainty; WildClawBench OpenClaw contributes as agent evidence where a model has an admitted result. The methodology page documents the weighting.
The newest WildClawBench OpenClaw publication ModelCap holds is dated 20 July 2026. The page re-renders every minute from the sealed dataset, so it reflects the latest refresh of the source board.