IFM K2-Horizon-375B Card - AutomationBench (Public): leaderboard
AutomationBench measures how well agents automate multi-step business workflows across connected tools. This series is the public task set as scored in IFM's K2-Horizon-375B-A23B model card, whose values differ from the Artificial Analysis and vendor-published AutomationBench numbers. Score in percent, higher is better; effort settings are in the model labels.
Metric: Score (%). Source: huggingface.co. Saturation forecast: Around September 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 5 (Max) | 34.7 | #65 (Claude Sonnet 5) |
| 2 | GPT-5.6 Luna (Max) | 33.5 | #79 (GPT-5.6 Luna) |
| 3 | GPT-5.6 Terra (High) | 28 | #40 (GPT-5.6 Terra) |
| 4 | GLM-5.2 (Max) | 26.2 | #80 (GLM-5.2) |
| 5 | MiniMax-M3 | 20.5 | #151 |
| 6 | Inkling (xHigh) | 12.8 | #144 (Inkling) |
| 7 | Nemotron 3 Ultra | 8 | #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-automationbench-public · How It Works · Data refreshed daily, snapshot 2026-11-05.