ClawMark - Red-Line Failure Rate: leaderboard
ClawMark (2026) runs coworker agents through 100 multi-day tasks in 13 professional scenarios on stateful sandboxed email, calendar, file, knowledge-base and spreadsheet services whose state changes between turns, scored by deterministic Python checkers; every model runs in the OpenClaw agent framework. This board is the share of the 55 red-line (must-not-violate) checks the agent fails, in percent; lower is better.
Metric: Red-line failure rate (%): share of the 55 red-line (must-not-violate) checker evaluations the agent fails; 100 multi-day coworker tasks in 13 professional scenarios on five stateful sandboxed services (filesystem, email, calendar, knowledge base, spreadsheet) with multimodal evidence; every model runs in the OpenClaw agent framework, one sweep, deterministic Python checkers (no LLM judge); lower is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 7 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 3.6 | #102 | |
| 2 | GPT-5.4 | High | 3.6 | #138 |
| 3 | Claude Sonnet 4.6 | Thinking | 3.6 | #179 |
| 4 | Claude Opus 4.6 | Thinking | 5.5 | #114 |
| 5 | Kimi K2.6 | 7.3 | #197 | |
| 6 | Kimi K2.5 | 9.1 | #272 | |
| 7 | Qwen 3.6 Plus | 14.5 | #245 |
No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=clawmark-red-line-failure-rate · How It Works · Data refreshed daily, snapshot 2026-11-22.