WildClawBench B
AI benchmark · Agentic
What is WildClawBench?
WildClawBench measures how well autonomous AI agents complete 60 hand-crafted real-world tasks inside a live OpenClaw environment with real shell, browser, email and files.
“Code Intelligence 12 tasks · where agents should shine, often don't”
internlm.github.io/WildClawBench/ ↗“Agents get access to a real bash shell, a real file system, a real browser, real email and calendar services.”
internlm.github.io/WildClawBench/ ↗“Productivity Flow 10 tasks · paper classification, scheduling, crawling”
internlm.github.io/WildClawBench/ ↗“Creative Synthesis 11 tasks · video editing, dubbing, cross-modal”
internlm.github.io/WildClawBench/ ↗“Safety Alignment 10 tasks · adversarial prompts, credential leaks”
internlm.github.io/WildClawBench/ ↗General capability
“Search & Retrieval 11 tasks · conflicting info, fuzzy matching”
internlm.github.io/WildClawBench/ ↗
Who built WildClawBench
Led by InternLM.
Tier B
Established: 15 or more models scored, or curated RL directories rate it.
How much stands behind a tier: usage, who backs it, a confirmed line from its own site or paper, and a clear read on X. High with three of the four, medium with two, low otherwise.
- 36 models scored
- in curated RL directories (1 pt)
On X · unclear, low buzz, from 2 posts about WildClawBench in the last month
Mentioned by Transluce for surfacing agents bypassing bot firewalls, and used by an indie developer to compare agent frameworks; little broader discussion.
How we got to tier B
- Adoption: 36 models scored. Tier B.
- On X: unclear, low buzz. “Mentioned by Transluce for surfacing agents bypassing bot firewalls, and used by an indie developer to compare agent frameworks; little broader discussion.” No move.
- In curated RL directories: 1 point, a floor of none.
- Confirmed: a line from its own site or paper places it.
A clear view on X moves a tier one step up or down; most entries don't move. How tiers work.
Where WildClawBench sits
Its domain on the map, live. Hover a logo for its name; click to open it. Open it on the full map
Next: browse all benchmarks. Entry from the RL Research daily scrape of public sources, 2026-11-14 snapshot.