Chat2Workflow - Developer: leaderboard
Chat2Workflow (2026) tests whether language models build executable Dify workflows from multi-turn natural-language requests drawn from real business workflows, run against test cases. Resolve rate in percent on the developer tasks; higher is better.
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 February 2027. 15 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 44.44 | #119 |
| 2 | GPT-5.2 | 38.89 | #185 |
| 3 | Claude Sonnet 4.5 | 31.94 | #261 |
| 4 | Kimi K2 (Thinking) | 26.39 | |
| 5 | Qwen 3 235B A22B | 20.83 | #628 |
| 6 | GLM-4.7 | 19.45 | #380 |
| 7 | Qwen 3 32B | 15.28 | #972 |
| 8 | Qwen 3 14B | 11.11 | #1269 |
| 9 | GPT-5.1 | 8.33 | #240 |
| 10 | Kimi K2 | 5.56 | #443 |
| 11 | GLM-4.6 | 5.55 | #498 |
| 12 | DeepSeek V3.2 | 4.17 | #419 |
| 13 | DeepSeek V3.1 | 2.78 | #511 |
| 14 | Qwen 3 8B | 1.39 | #1873 |
| 15 | Qwen 3 Coder 480B A35B Instruct | 0 | #652 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=chat2workflow-developer · How It Works · Data refreshed daily, snapshot 2026-11-21.