PeakBench - Logical Planning: leaderboard
Dependency-edge F1 for tool-invocation plans in PeakBench. Scores range from zero to one and measure logical planning separately from execution scheduling.
Metric: Dependency Edge F1 (0-1). Source: arxiv.org. Saturation forecast: Around January 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 0.84 | #98 |
| 2 | DeepSeek V4 Pro | 0.81 | #102 |
| 3 | GLM-5 | 0.77 | #141 |
| 4 | O3 | 0.77 | #124 |
| 5 | Claude Sonnet 4.6 | 0.76 | #90 |
| 6 | DeepSeek V4 Flash | 0.73 | #137 |
| 7 | Kimi K2.5 | 0.69 | #144 |
| 8 | GPT-4.1 | 0.66 | #248 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=peakbench-logical-planning · How It Works · Data refreshed daily, snapshot 2026-10-20.