Benchmark
NL2Repo Benchmark
What is NL2Repo?
NL2Repo evaluates long-horizon coding capabilities including repository-level understanding, where models must generate or modify code across entire…
- Modality
- text
- Evaluates
- agents, code
- Openness
- unknown
- Importer
- LLM Stats
- Reported scores
- 17
Source provenance
- Original evidence https://api.zeroeval.com/leaderboard/benchmarks/nl2repo?top_n=500
NL2Repo results and reported scores
LLM Stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| DeepSeek-V4-Flash-0731 | DeepSeek | 0.542 | 2026-07-31 | evidence |
| DeepSeek-V4-Pro-0813 | DeepSeek | 0.615 | 2026-08-13 | evidence |
| GLM-5.1 | Z.ai | 0.427 | 2026-04-07 | evidence |
| GLM-5.2 | Z.ai | 0.489 | 2026-06-16 | evidence |
| GLM-5.3 | Z.ai | 0.58 | 2026-08-14 | evidence |
| Hy3 | Tencent | 0.456 | 2026-07-06 | evidence |
| MiniMax M2.7 | MiniMax | 0.398 | 2026-03-18 | evidence |
| MiniMax M3 | MiniMax | 0.4213 | 2026-06-01 | evidence |
| Qwen3.6 Plus | Qwen | 0.379 | 2026-04-02 | evidence |
| Qwen3.6-27B | Qwen | 0.362 | 2026-04-21 | evidence |
| Qwen3.6-35B-A3B | Qwen | 0.294 | 2026-04-16 | evidence |
| Qwen3.7 Max | Qwen | 0.472 | 2026-05-19 | evidence |
| Qwen3.7-Plus | Qwen | 0.411 | 2026-05-31 | evidence |
| Qwen3.8 Max | Qwen | 0.559 | 2026-08-02 | evidence |
| Qwen3.8-27B | Qwen | 0.423 | 2026-08-14 | evidence |
| Seed 2.1 Pro | ByteDance | 0.47 | 2026-06-24 | evidence |
| Seed 2.1 Turbo | ByteDance | 0.437 | 2026-06-24 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.