IFM K2-Horizon-375B Card - WildClawBench (English Text-Only Subset): leaderboard
WildClawBench tests agents on in-the-wild agentic tasks. IFM's K2-Horizon-375B-A23B model card runs a subset of its English text-only tasks, so values are not comparable with full-benchmark results. Score in percent, higher is better; Claude Sonnet 5 has no cell and effort settings are in the model labels.
Metric: Score (%). Source: huggingface.co. Saturation forecast: Around August 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Terra (High) | 60 | #40 (GPT-5.6 Terra) |
| 2 | MiniMax-M3 | 56.4 | #151 |
| 3 | GLM-5.2 (Max) | 55 | #80 (GLM-5.2) |
| 4 | Inkling (xHigh) | 52.3 | #144 (Inkling) |
| 5 | GPT-5.6 Luna (Max) | 50.4 | #79 (GPT-5.6 Luna) |
| 6 | Nemotron 3 Ultra | 34.2 | #245 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=ifm-k2-horizon-375b-card-wildclawbench-english-text-only-subset · How It Works · Data refreshed daily, snapshot 2026-11-05.