APIFlow-Bench 1.0 - Solo Subtasks: leaderboard
Metric: Pass Rate (%). Source: www.postman.com. 24 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K3 | 97.84 |
| 2 | Claude Sonnet 5 | 96.51 |
| 3 | Claude Opus 4.8 | 96.02 |
| 4 | Claude Sonnet 4.6 | 95.68 |
| 5 | DeepSeek V4 Flash | 95.6 |
| 6 | Kimi K2.7 Code | 95.44 |
| 7 | Kimi K2.6 | 95.19 |
| 8 | Qwen 3.6 Plus | 95.1 |
| 9 | GPT-5.5 | 94.94 |
| 10 | DeepSeek V4 Pro | 94.61 |
| 11 | Kimi K2.5 | 94.36 |
| 12 | GPT-5.6 Sol | 94.36 |
| 13 | GLM-5.2 | 93.86 |
| 14 | Qwen 3.7 Plus | 93.53 |
| 15 | GPT-5.6 Luna | 93.36 |
Interactive version: theaggregate.ai/benchmark?slug=apiflow-bench-1-0-solo-subtasks · How It Works · Data refreshed daily, snapshot 2026-09-19.