APIFlow-Bench 1.0 - Solo Subtasks: leaderboard
Postman's APIFlow-Bench 1.0 measures whether agents complete long, dependent REST API workflows (arXiv 2608.29128v1). This board covers the solo set of 241 single-subtask tasks; pass rate is passed over all trials, 0-100, higher is better, and refusals and abstentions count as failures. From the live leaderboard's overview-solo page.
Metric: Pass Rate (%). Source: www.postman.com. Saturation forecast: Around February 2027. 24 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Kimi K3 | 97.84 | #35 |
| 2 | Claude Sonnet 5 | 96.51 | #65 |
| 3 | Claude Opus 4.8 | 96.02 | #39 |
| 4 | Claude Sonnet 4.6 | 95.68 | #94 |
| 5 | DeepSeek V4 Flash | 95.6 | #155 |
| 6 | Kimi K2.7 Code | 95.44 | #95 |
| 7 | Kimi K2.6 | 95.19 | #113 |
| 8 | Qwen 3.6 Plus | 95.1 | #124 |
| 9 | GPT-5.5 | 94.94 | #33 |
| 10 | DeepSeek V4 Pro | 94.61 | #112 |
| 11 | Kimi K2.5 | 94.36 | #152 |
| 12 | GPT-5.6 Sol | 94.36 | #22 |
| 13 | GLM-5.2 | 93.86 | #80 |
| 14 | Qwen 3.7 Plus | 93.53 | #98 |
| 15 | GPT-5.6 Luna | 93.36 | #79 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=apiflow-bench-1-0-solo-subtasks · How It Works · Data refreshed daily, snapshot 2026-11-05.