ParallelKernelBench — leaderboard
Metric: Fast1@3 (% of problems). Source: www.together.ai. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 31.03 |
| 2 | Claude Opus 4.7 | 22.99 |
| 3 | Gemini 3 Pro | 21.84 |
| 4 | GLM-5.1 | 3.45 |
| 5 | GLM-5.2 | 2.3 |
| 6 | DeepSeek V4 Pro | 1.15 |
Interactive version: theaggregate.ai/benchmark?slug=parallelkernelbench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.