APIFlow-Bench 1.0 - Discovery: leaderboard
Postman's APIFlow-Bench 1.0 pass rate on the tasks that exercise the discovery capability axis, across all task lengths (arXiv 2608.29128v1). Passed over all trials, 0-100 to one decimal, higher is better; refusals and abstentions count as failures. From the live leaderboard's axis page in its all-tasks view.
Metric: Pass Rate (%). Source: www.postman.com. Saturation forecast: Not forecast. 24 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 82.5 | #94 |
| 2 | Claude Opus 4.8 | 82.5 | #39 |
| 3 | GLM-5.2 | 79.5 | #80 |
| 4 | GPT-5.5 | 78.9 | #33 |
| 5 | DeepSeek V4 Flash | 78.6 | #155 |
| 6 | Kimi K2.5 | 77.8 | #152 |
| 7 | Qwen 3.6 Plus | 77.8 | #124 |
| 8 | Kimi K3 | 77.8 | #35 |
| 9 | DeepSeek V4 Pro | 77.5 | #112 |
| 10 | GPT-5.6 Sol | 77.5 | #22 |
| 11 | Kimi K2.6 | 77.5 | #113 |
| 12 | GPT-5.6 Luna | 77.5 | #79 |
| 13 | Claude Haiku 4.5 (20251001) | 77.3 | #264 |
| 14 | Kimi K2.7 Code | 77.3 | #95 |
| 15 | MiniMax-M2.7 | 76.7 | #270 |
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-discovery · How It Works · Data refreshed daily, snapshot 2026-11-05.