Chat2Workflow - Developer: leaderboard
Metric: Resolve rate (%) on Chat2Workflow developer tasks: share of test cases where the workflow the model generates for the current round of a multi-turn request, deployed on Dify 1.9.2, executes and its output satisfies every requirement of the instruction, judged by GPT-5.1; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 44.44 |
| 2 | GPT-5.2 | 38.89 |
| 3 | Claude Sonnet 4.5 | 31.94 |
| 4 | Kimi K2 (Thinking) | 26.39 |
| 5 | Qwen 3 235B A22B | 20.83 |
| 6 | GLM-4.7 | 19.45 |
| 7 | Qwen 3 32B | 15.28 |
| 8 | Qwen 3 14B | 11.11 |
| 9 | GPT-5.1 | 8.33 |
| 10 | Kimi K2 | 5.56 |
| 11 | GLM-4.6 | 5.55 |
| 12 | DeepSeek V3.2 | 4.17 |
| 13 | DeepSeek V3.1 | 2.78 |
| 14 | Qwen 3 8B | 1.39 |
| 15 | Qwen 3 Coder 480B A35B Instruct | 0 |
Interactive version: theaggregate.ai/benchmark?slug=chat2workflow-developer · How It Works · Data refreshed daily, snapshot 2026-10-07.