DevBench Implementation
Models implement repository code. The score is the acceptance-test pass rate, with unit-test pass rate as an auxiliary metric.
One track of DevBench. The DevBench record keeps lab scores that combine tracks.
- Made by
- OpenCompass (Shanghai AI Laboratory) and collaborators
- Released
- March 2024
- Kind
- Building whole projects
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Agent task
- Tasks
- All public
- Strongest model tested
- GPT-4 Turbo Nov 2023 #175 on the capability index
- Last model added
- January 2024
- Scores refreshed
- November 12th, 2026
GPT-4 Turbo (Jan 2024) leads with 7.10% among models released by January 2024
Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.
The newest model with a score here came out in January 2024, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- Claude Fable 5.2
-
Gemini 4 Carbon - GPT-6 Astra
- Claude Opus 5.5
- and 16 more
- 1 GPT-4 Turbo (Jan 2024) 7.10%
- 2 GPT-4 Turbo Nov 2023 6.90%
- 3 DeepSeek-Coder-33B-Instruct 4.40%
- 4 GPT-3.5-Turbo (snapshot unspecified) 4.20%
- 5 CodeLlama-13B-Instruct 0.60%
- 6
Code Llama 34B Instruct 0.60% - 7 CodeLlama-7B-Instruct 0%
- 8 DeepSeek-Coder-1.3B-Instruct 0%
Source: Official DevBench Implementation leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4 Turbo (Jan 2024) | 7.10% | Jan 2024 | Official leaderboard | |
| 2 | GPT-4 Turbo Nov 2023 | 6.90% | Nov 2023 | Official leaderboard | |
| 3 | DeepSeek-Coder-33B-Instruct | 4.40% | Official leaderboard | ||
| 4 | GPT-3.5-Turbo (snapshot unspecified) | 4.20% | Official leaderboard | ||
| 5 | CodeLlama-13B-Instruct | 0.60% | Official leaderboard | ||
| 6 | Code Llama 34B Instruct | 0.60% | Aug 2023 | Official leaderboard | |
| 7 | CodeLlama-7B-Instruct | 0% | Official leaderboard | ||
| 8 | DeepSeek-Coder-1.3B-Instruct | 0% | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official DevBench Implementation 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 DevBench Implementation leaderboard
No lab model card reports it yet
We read the headline results of 522 launch posts from 52 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