OrchestrationBench: leaderboard
Multi-step workflow orchestration benchmark testing planning, function calling accuracy, and call rejection ability in Korean and English.
Metric: Average Score. Source: github.com. Saturation forecast: Around January 2027. 17 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 85.07 | #49 |
| 2 | Gemini 3 Pro (Preview) | 84.12 | #66 |
| 3 | Gemini 3 Flash (Preview) | 83.54 | #82 |
| 4 | GLM-4.7 | 83.17 | #192 |
| 5 | Gemini 3.1 Pro (Preview) | 82.8 | #58 |
| 6 | GPT-5.4 (2026-03-05) | 78.15 | #59 |
| 7 | GPT-5.2 | 76.05 | #110 |
| 8 | Qwen 3 235B A22B 2507 Instruct | 70.65 | #300 |
| 9 | Solar Open 100B | 67.05 | #406 |
| 10 | Qwen 3 30B A3B 2507 Instruct | 66.08 | #488 |
| 11 | Qwen 2.5 32B Instruct | 66.03 | #498 |
| 12 | A.X 4.0 | 65.98 | #424 |
| 13 | K-EXAONE | 58.3 | #378 |
| 14 | HyperCLOVA X SEED Think (32B) | 54.36 | #663 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=orchestrationbench · How It Works · Data refreshed daily, snapshot 2026-10-20.