EnterpriseClawBench (Hermes): leaderboard
FrontisAI's EnterpriseClawBench — enterprise software-engineering agent tasks run with the Hermes agent scaffold; primary resolved score (%) over a small frontier pool.
Metric: Primary score (%). Source: frontisai.github.io. Saturation forecast: Around January 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 | 61.96 | #31 |
| 2 | DeepSeek V4 Pro | 58.55 | #102 |
| 3 | Kimi K2.6 | 56.13 | #104 |
| 4 | MiniMax-M3 | 52.46 | #136 |
| 5 | Claude Sonnet 4.6 | 45.75 | #90 |
| 6 | Claude Opus 4.6 | 42.54 | #63 |
| 7 | Claude Haiku 4.5 | 40.48 | #280 |
| 8 | GPT-4.1 Mini | 28.59 | #355 |
| 9 | Qwen 3 235B A22B | 27.03 | #313 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=enterpriseclawbench-hermes · How It Works · Data refreshed daily, snapshot 2026-10-20.