Phi-Bench - Long-Horizon Implementation: leaderboard
Phi-Bench evaluates agents implementing and optimizing AI infrastructure, including kernels, serving, compression and system engineering. Scores combine correctness and performance under the paper's shared harness; task formats stay separate. This series reports Phi-Bench - Long-Horizon Implementation from 2609.10226v1, Sx5.T3: Infrastructure Score (%). Higher is better.
Metric: Infrastructure Score (%). Source: arxiv.org. Saturation forecast: Around 2029. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 | 21.6 | #19 |
| 2 | Kimi K3 | 19.55 | #53 |
| 3 | Qwen 3.8 Max | 16.61 | #56 |
| 4 | GPT-5.6 Sol | 13.98 | #29 |
| 5 | GLM-5.2 | 13.65 | #138 |
| 6 | Claude Sonnet 5 | 13.46 | #108 |
| 7 | Qwen 3.7 Max | 12.79 | #143 |
| 8 | DeepSeek V4 Pro | 11.45 | #188 |
No result here: #4 Claude Opus 5.5, #5 GPT-6 Astra, #9 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=phi-bench-long-horizon-implementation · How It Works · Data refreshed daily, snapshot 2026-11-08.