RepoNormBench
Agents change code in existing repositories while following the contribution rules that apply to each task.
The board reports functional success and rule compliance separately for 121 tasks and three coding models.
- Made by
- SYSUSELab
- Released
- October 2026
- Kind
- Fixing real repositories
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Terminal
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Agents
- Strongest model tested
-
DeepSeek-V4-Flash #87 on the capability index
- Last model added
- April 2026
- Scores refreshed
- November 20th, 2026
DeepSeek-V4-Flash leads with 68.6% among models released by April 2026
Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. The board lists functional success first and reports norm compliance separately. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.
The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
Source: Official RepoNormBench leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 68.6% | RepoNorm (Accepted-Only), mini_swe_agent_v2 | Apr 2026 | Official leaderboard |
| 2 | | 52.1% | RepoNorm (Accepted-Only), mini_swe_agent_v2 | Feb 2026 | Official leaderboard |
| 3 | | 37.2% | CodeWiki, mini_swe_agent_v2 | Mar 2026 | Official leaderboard |
Rankings by part
We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Overall NCR, Contribution NCR and Prompt-omitted NCR. DeepSeek-V4-Flash leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
Overall NCR
3 models we track have a score here.
- 1
DeepSeek-V4-Flash 72.6%
- 2
Qwen3.5-Flash 68.7%
- 3
Gemini 3.1 Flash-Lite 63.4%
Contribution NCR
3 models we track have a score here.
- 1
DeepSeek-V4-Flash 49.2%
- 2
Qwen3.5-Flash 43.4%
- 3
Gemini 3.1 Flash-Lite 33.3%
Prompt-omitted NCR
3 models we track have a score here.
- 1
DeepSeek-V4-Flash 61.3%
- 2
Qwen3.5-Flash 55.8%
- 3
Gemini 3.1 Flash-Lite 49.7%
Functional + all norms
3 models we track have a score here.
- 1
DeepSeek-V4-Flash 8.26%
- 2
Qwen3.5-Flash 4.96%
- 3
Gemini 3.1 Flash-Lite 1.65%
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: Official RepoNormBench leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| RepoNorm (Accepted-Only) | mini_swe_agent_v2 | 68.6% | |||
| CodeWiki | mini_swe_agent_v2 | 67.8% | |||
| Raw | mini_swe_agent_v2 | 59.5% | |||
| RepoNorm (Accepted-Only) | mini_swe_agent_v2 | 52.1% | |||
| Raw | mini_swe_agent_v2 | 44.6% | |||
| CodeWiki | mini_swe_agent_v2 | 44.6% | |||
| CodeWiki | mini_swe_agent_v2 | 37.2% | |||
| RepoNorm (Accepted-Only) | mini_swe_agent_v2 | 36.4% | |||
| Raw | mini_swe_agent_v2 | 30.6% |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official RepoNormBench leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official RepoNormBench leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Coding
- SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
- LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests.
DeepSeek-V4-Pro 93.5%
- SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
- DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
- SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases.
Seed2.1 Pro 59.8%
- Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance.
DeepSeek V4.1 Flash 3,471
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for DeepSeek-V4-Flash. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (3 entries).
- Note Models scored Only 3 models we track.
- Warning Newest score The newest came 7 months ago, in April 2026, for DeepSeek-V4-Flash.
- Note Frontier models None of the top 10 has a score. The strongest tested is DeepSeek-V4-Flash (#87).
- Pass Headroom The best is 68.6%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on huggingface.co: 9 rows for models we track at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.