IDE-Bench: leaderboard
IDE coding-agent benchmark with 80 interactive software-engineering tasks that test whether models can make a correct fix on the first attempt.
Metric: Pass@1 Accuracy (self-reported). Source: benchmarklist.com. Status: saturation imminent. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 | 87.5 |
| 2 | GPT-5.2 | 85 |
| 3 | Claude Opus 4.5 | 83.75 |
| 4 | Claude Haiku 4.5 | 78.75 |
| 5 | GPT-5.1 Codex Max | 73.75 |
| 6 | Qwen 3 Max | 65 |
| 7 | Gemini 3 Pro | 55 |
| 8 | Grok 4.1 Fast | 35 |
| 9 | DeepSeek V3 | 31.25 |
| 10 | DeepSeek R1 | 20 |
| 11 | Grok Code Fast 1 | 11.25 |
| 12 | Llama 4 Maverick | 2.5 |
| 13 | Llama 4 Scout | 2.5 |
| 14 | Command-R+ | 0 |
Interactive version: theaggregate.ai/benchmark?slug=ide-bench · How It Works · Data refreshed daily, snapshot 2026-09-05.