DiagFlowBench — leaderboard
Metric: Step Accuracy (self-reported). Source: benchmarklist.com. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.3 70B Instruct | 85 |
| 2 | Gemini 2.5 Flash | 84 |
| 3 | Llama 4 Maverick | 82.8 |
| 4 | Llama 4 Scout | 79.2 |
| 5 | GPT-OSS-120B | 76.2 |
| 6 | GPT-4o Mini | 75.4 |
| 7 | Qwen 3 235B A22B | 73.7 |
| 8 | Nemotron 3 Super | 70.3 |
| 9 | Qwen 3 30B A3B | 70.1 |
Interactive version: theaggregate.ai/benchmark?slug=diagflowbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.