DevOps-Gym
Agents build, monitor, repair and test Java and Go projects. The Prime port converts tasks to single-turn questions with proxy grading.
The January 2026 paper compares Claude, Gemini and DeepSeek. Claude Code led the build and configuration comparison at 58.33%.
- Made by
- UCSB ML Security Lab and collaborators
- Released
- January 2026
- Kind
- Fixing real repositories
- Level
- University
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Terminal
- Items from
- Another benchmark
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Agents
- Strongest model tested
- Claude Sonnet 4 #113 on the capability index
- Last model added
- May 2025
- Scores refreshed
- November 20th, 2026
SageAgent + GPT-5.3 Codex tops the leaderboard with 46.8% Claude Sonnet 4 leads with 27.5% Claude Sonnet 4 leads with 9.67% in mini-SWE-agent
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran on its own with several settings, we keep its best. Agents other companies built around a model get a pale shade of that model's lab color. Gray marks blends and entries not in our model list, such as fine-tunes and older models. None of these count toward leaders. The best model on its own is Claude Sonnet 4, at 27.5%. Agents built around a model, such as SageAgent + GPT-5.3 Codex, are their own entries. We don't count them as the model's score. This view shows only models we track, each at its best. This view keeps only runs in mini-SWE-agent, a harness no model’s maker built. A maker’s own agent, such as Claude Code with a Claude model, is left out.
- 1 SageAgent + GPT-5.3 Codex 46.8%
- 2 Claude Sonnet 4 27.5%
- 3 OpenHands + Claude Sonnet 4 21%
- 4 OpenHands + o4-mini 13%
- 5 OpenHands + Qwen3-Coder-30B-A3B-Instruct 11.4%
- 6 Gemini-2.5-Pro (snapshot unspecified) 10.6%
- 7 OpenHands + DeepSeek-V3.1 7.13%
- 8 Aider + Claude Sonnet 4 4.37%
Source: Official DevOps-Gym leaderboard
- 1 Claude Sonnet 4 27.5%
Source: Official DevOps-Gym leaderboard
- 1 Claude Sonnet 4 9.67%
Source: Official DevOps-Gym leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | SageAgent + GPT-5.3 Codex | 46.8% | Feb 2026 | Official leaderboard | |
| 2 | Claude Sonnet 4 | 27.5% | Claude Code | May 2025 | Official leaderboard |
| 3 | OpenHands + Claude Sonnet 4 | 21% | May 2025 | Official leaderboard | |
| 4 | OpenHands + o4-mini | 13% | Apr 2025 | Official leaderboard | |
| 5 | OpenHands + Qwen3-Coder-30B-A3B-Instruct | 11.4% | Jul 2025 | Official leaderboard | |
| 6 | Gemini-2.5-Pro (snapshot unspecified) | 10.6% | OpenHands | Official leaderboard | |
| 7 | OpenHands + DeepSeek-V3.1 | 7.13% | Aug 2025 | Official leaderboard | |
| 8 | Aider + Claude Sonnet 4 | 4.37% | May 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4 | 27.5% | Claude Code | May 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4 | 9.67% | mini-SWE-Agent | May 2025 | Official 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 DevOps-Gym leaderboard
Source: Official DevOps-Gym 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
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 18 months ago, in May 2025, for Claude Sonnet 4. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the official leaderboard (8 entries).
- Pass Models scored 1 model we track, plus 7 other entries on its leaderboard.
- Warning Newest score The newest came 18 months ago, in May 2025, for Claude Sonnet 4.
- Note Frontier models None of the top 10 has a score. The strongest tested is Claude Sonnet 4 (#113).
- Pass Headroom The best is 27.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on devops-gym.com: 8 rows for models we track and 1 other entry 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.