Explicit Edit Benchmark: leaderboard
226 deterministic exact-edit tasks over eleven families - block, cross-file, distinct, multi-file and subset edits, replace-all, select-one, unique-match, literal edge cases, Unicode fixes and non-Latin-script text - replayed through coding agents; an observation passes only when the edited file matches the expected bytes. Score v2 = coverage × (0.75 first-exact + 0.25 final-exact), repetitions averaged inside a configuration and task and configurations weighted equally inside a task. Published here per model family, upstream's median over that model's complete eligible configurations, so a model with many harnesses is not ranked by its luckiest one. Higher is better. Source: huggingface.co/datasets/alexshpunt/explicit-edit-benchmark.
Metric: Explicit Edit Score v2 (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 26 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6.1 Sol | 100 | #7 |
| 2 | GPT-6.1 Terra | 100 | #15 |
| 3 | DeepSeek V4 Pro | 99.7 | #112 |
| 4 | Gemini 3.8 Flash | 99.7 | #30 |
| 5 | GPT-5.6 Sol | 98.8 | #22 |
| 6 | DeepSeek V4.1 Flash | 98.7 | #37 |
| 7 | MiMo-V2.6-Pro | 98.6 | #48 |
| 8 | DeepSeek V4 Flash | 97.9 | #155 |
| 9 | Qwen 3.8 Flash | 97.2 | #62 |
| 10 | GPT-6 Sol | 97.1 | #19 |
| 11 | Grok 4.6 | 97.1 | #29 |
| 12 | Kimi K2.7 Code | 96.7 | #95 |
| 13 | MiMo-V2.6-Flash | 96.7 | #74 |
| 14 | GPT-5.6 Terra | 96.3 | #40 |
| 15 | GPT-5.6 Luna | 94.1 | #79 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=explicit-edit-benchmark · How It Works · Data refreshed daily, snapshot 2026-11-05.