Chat2Workflow - AIGC: 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 AI-generated content tasks; higher is better.
Metric: Resolve rate (%) on Chat2Workflow AI-generated content 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) | 54.32 | #119 |
| 2 | GLM-4.7 | 50.61 | #380 |
| 3 | GPT-5.1 | 45.06 | #240 |
| 4 | Kimi K2 (Thinking) | 43.83 | |
| 5 | Claude Sonnet 4.5 | 41.97 | #261 |
| 6 | DeepSeek V3.1 | 40.12 | #511 |
| 7 | DeepSeek V3.2 | 39.51 | #419 |
| 8 | GLM-4.6 | 39.51 | #498 |
| 9 | Kimi K2 | 35.19 | #443 |
| 10 | GPT-5.2 | 32.1 | #185 |
| 11 | Qwen 3 32B | 29.63 | #972 |
| 12 | Qwen 3 235B A22B | 23.46 | #628 |
| 13 | Qwen 3 Coder 480B A35B Instruct | 23.46 | #652 |
| 14 | Qwen 3 14B | 22.22 | #1269 |
| 15 | Qwen 3 8B | 6.79 | #1873 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=chat2workflow-aigc · How It Works · Data refreshed daily, snapshot 2026-11-21.