RealBench (RAG) - Level 1: leaderboard
Metric: Test pass rate (%), passed tests over all tests of each task, averaged over the level-1 tasks (14 repositories of about 260 lines), with retrieval-augmented generation (file by file, each conditioned on the earlier generated files it imports); RealBench repository-level Python code generation from natural-language requirements plus UML package and class diagrams (61 real repositories in four size levels), one greedy solution per task through the official APIs; higher is better. Source: arxiv.org. Saturation forecast: Around September 2027. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V3 | 41.85 |
| 2 | Claude Sonnet 4 (20250514) | 26.52 |
| 3 | GPT-4o (2024-05-13) | 24.6 |
| 4 | Gemini 2.5 Flash | 19.4 |
| 5 | Qwen 3 235B A22B | 9.23 |
| 6 | Qwen 2.5 Coder 7B Instruct | 8.95 |
Interactive version: theaggregate.ai/benchmark?slug=realbench-rag-level-1 · How It Works · Data refreshed daily, snapshot 2026-10-07.