PatchEval - Location Oracle: leaderboard
Metric: % POC & Unit. Source: patcheval.github.io. Saturation forecast: Around 2030. 26 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 29.13 |
| 2 | Gemini 2.5 Pro | 22.61 |
| 3 | DeepSeek V3.2 Speciale | 21.74 |
| 4 | GPT-4.1 | 21.3 |
| 5 | DeepSeek R1 | 19.57 |
| 6 | DeepSeek R1 0528 | 19.13 |
| 7 | GPT-4.1 (2025-04-14) | 18.7 |
| 8 | DeepSeek V3.2 (Thinking) | 18.7 |
| 9 | DeepSeek V3 (0324) | 17.83 |
| 10 | Qwen 3 Max | 17.83 |
| 11 | Qwen 3 Coder 480B A35B Instruct | 17.83 |
| 12 | O3 (2025-04-16) | 17.39 |
| 13 | Kimi K2 | 16.52 |
Interactive version: theaggregate.ai/benchmark?slug=patcheval-location-oracle · How It Works · Data refreshed daily, snapshot 2026-10-03.