ComplexFuncBench: leaderboard
1,000 multi-step function-calling tasks over 43 live travel APIs (hotels, flights, attractions, car rental, taxi) with 128k contexts, from Zhipu AI and Tsinghua (2025); success rate, GPT-4o 60.5%.
Metric: Score (imported). Source: arxiv.org. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 20 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o (2024-11-20) | 66.5 | #381 |
| 2 | GPT-4.1 | 65.5 | #248 |
| 3 | GPT-4.5 | 63 | #274 |
| 4 | Claude 3.5 Sonnet | 61 | #350 |
| 5 | GPT-4o (2024-08-06) | 60.5 | #336 |
| 6 | GPT-4 Turbo | 49.5 | #404 |
| 7 | GPT-4.1 Mini | 49.3 | #355 |
| 8 | O1 (High) | 47.6 | #183 (O1) |
| 9 | Claude 3.5 Haiku | 45.8 | #568 |
| 10 | Qwen 2.5 72B | 40.1 | #499 |
| 11 | GPT-4o Mini | 38.6 | #599 |
| 12 | Mistral Large 2 (Jul) | 20.1 | #484 |
| 13 | O3 Mini (High) | 17.6 | #272 (O3 Mini) |
| 14 | glm-4-9B | 9.4 | #889 |
| 15 | GPT-4.1 Nano | 5.7 | #731 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=complexfuncbench · How It Works · Data refreshed daily, snapshot 2026-10-20.