Benchmark Almanac

SWE-rebench V2

Claude Opus 4.5 25.2% leads since February 2026
Not in model cards
Version 2

Agents repair real repository issues using executable tests. The authors compare models on 300 tasks across five languages. Harbor also ports 51 validated Clojure tasks.

In June 2026, Claude Opus 4.5 led seven models on the 300-task sample with 25.2% pass@1. The Clojure port has no published model scores.

Made by
Nebius
Released
February 2026
Kind
Fixing real repositories
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Code → Code
Interaction
Agentic
Environment
Terminal
Items from
Real work and use
Run by
Anyone
Based on
SWE-rebench
In views
Agents
Strongest model tested
GPT-5.2 #51 on the capability index
Last model added
December 2025
Scores refreshed
November 20th, 2026
7
Models scored in its paper

Claude Opus 4.5 leads with 25.2% among models released by December 2025

Best score per model from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in December 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Claude Opus 4.5 25.2%
  2. 2 GLM-4.7 21.3%
  3. 3 MiniMax-M2.1 19.2%
  4. 4 Gemini 3 Flash 18.1%
  5. 5 DeepSeek-V3.2 17.4%
  6. 6 GPT-5.2 17%
  7. 7 gpt-oss-120b 8.80%

Source: The SWE-rebench V2 paper

#ModelScoreSettingDateSource
1 Claude Opus 4.5 25.2% Nov 2025 Official leaderboard
2 GLM-4.7 21.3% Dec 2025 Official leaderboard
3 MiniMax-M2.1 19.2% Dec 2025 Official leaderboard
4 Gemini 3 Flash 18.1% Dec 2025 Official leaderboard
5 DeepSeek-V3.2 17.4% Dec 2025 Official leaderboard
6 GPT-5.2 17% Dec 2025 Official leaderboard
7 gpt-oss-120b 8.80% Aug 2025 Official leaderboard

Rankings by part

We read 5 parts from the benchmark’s paper, each ranking models on one part of the test, such as Python, JavaScript and Go. Claude Opus 4.5 leads 4 of them, and GLM-4.7 leads the other one. Arrows show how far a model moved from its overall rank among the models we track.

Python

7 models we track have a score here.

  1. 1 Claude Opus 4.5 36.1%
  2. 2 GLM-4.7 27.2%
  3. 3 MiniMax-M2.1 26.1%
  4. 4 Gemini 3 Flash 25.6%
  5. 5 DeepSeek-V3.2 23.3%
  6. 6 GPT-5.2 20.6%
  7. 7 gpt-oss-120b 8.90%

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic
Benchmark paper Not tested since Feb 2026

Source: The SWE-rebench V2 paper

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.5 gpt-oss-120b 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.5 gpt-oss-120b

Source: Official SWE-rebench V2 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.

Versions

We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.

Qwen Xiaomi Corp MiniMax Anthropic
SWE-rebench V2 Not tested since Feb 2026
SWE-rebench Not tested since Jul 2026
SWE-rebench V2 Not tested since Feb 2026

Source: Official leaderboards

More in Coding

All 523 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 11 months ago, in December 2025, for MiniMax-M2.1. 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 benchmark’s paper (7 entries).
  • Pass Models scored 7 models we track.
  • Warning Newest score The newest came 11 months ago, in December 2025, for MiniMax-M2.1.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.2 (#51).
  • Pass Headroom The best is 25.2%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 7 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.

How we check