SupChain-Bench (Tool Calling, With SOP) - Level 3: leaderboard
Metric: Information retrieval accuracy (%) on the 35 Level 3 questions (over 25 steps, up to 7 distinct tools) of SupChain-Bench's order-diagnosis tool-calling questions; the prompt adds the expert standard operating procedure, step by step; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 88.6 | #91 |
| 2 | Claude Sonnet 4 | 28.6 | #194 |
| 3 | Gemini 2.5 Pro | 25.7 | #145 |
| 4 | Qwen 3 Max | 17.1 | #201 |
| 5 | DeepSeek V3 | 11.4 | #312 |
| 6 | Kimi K2 | 5.7 | #236 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=supchain-bench-tool-calling-with-sop-level-3 · How It Works · Data refreshed daily, snapshot 2026-10-11.