RealBench (Holistic) - Level 4: leaderboard

RealBench (2026) asks LLMs to build a whole Python repository from natural-language requirements and UML system designs, here with holistic generation (the whole repository in one pass) on the level-4 repositories (about 2,360 lines of code). The score is the percentage of the original unit tests that the generated repository passes, higher is better.

Metric: Test pass rate (%), passed tests over all tests of each task, averaged over the level-4 tasks (10 repositories of about 2,360 lines), with holistic generation (the whole repository in one pass); 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 December 2027. 6 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Claude Sonnet 4 (20250514)14.3#410
2GPT-4o (2024-05-13)13.51#740
3Gemini 2.5 Flash3.07#466
4DeepSeek V32.21#736
5Qwen 3 235B A22B0.91#633
6Qwen 2.5 Coder 7B Instruct0.85#3976

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=realbench-holistic-level-4 · How It Works · Data refreshed daily, snapshot 2026-11-22.