AccelerateBench v1.1: leaderboard
HPC and scientific-computing program optimization benchmark (arXiv:2611.20882v1, Table 1) evaluating 16 frontier coding agents (at High reasoning effort in Claude Code or Codex) across 32 tasks from six supercomputing competitions. Scored by Score@1, the mean fraction of baseline runtime saved (%) on correct submissions.
Metric: Score@1 (mean runtime savings %). Source: arxiv.org. Saturation forecast: Around June 2028. 16 models tracked. Updated 21 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Claude Sonnet 5.5 | High | 67.4 | #24 |
| 2 | Claude Opus 5.5 | High | 59.4 | #9 |
| 3 | GPT-6.1 Sol | High | 49.2 | #15 |
| 4 | GPT-6 Astra | High | 47.2 | #10 |
| 5 | Hy4 preview | High | 45.2 | #74 |
| 6 | hy4 | High | 45 | #40 |
| 7 | Kimi K3 | High | 42.5 | #61 |
| 8 | DeepSeek V4.1 Flash (1017) | High | 40.1 | #47 |
| 9 | GLM-5.3 | High | 39.7 | #72 |
| 10 | GLM-5.3 Flash | High | 39.3 | #133 |
| 11 | DeepSeek V4 Pro | High | 37.9 | #196 |
| 12 | GPT-6.1 Terra | High | 37.9 | #25 |
| 13 | DeepSeek V4 Flash | High | 34.1 | #267 |
| 14 | GLM-5.2 | High | 34 | #150 |
| 15 | Grok 4.6 | High | 33.9 | #50 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=acceleratebench-v1-1 · How It Works · Data refreshed daily, snapshot 2026-11-21.