Benchmark Almanac

RH SWE-bench v1

Claude Opus 4.6 63.3% leads since July 2026
Not in model cards
Version 1

Agents repair software across Red Hat ecosystem repositories and test each change against the issue requirements.

The original version contains 357 tasks across 25 repositories. Red Hat reduced the revised version to 337 tasks.

Released
July 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, Actions
Interaction
Agentic
Environment
Terminal
Items from
Real work and use
Run by
Anyone
In views
Agents
Strongest model tested
Claude Opus 4.6 #39 on the capability index
Last model added
April 2026
Scores refreshed
November 20th, 2026
5
Models scored on its own leaderboard

Claude Opus 4.6 leads with 63.3% among models released by April 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. 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:

  1. 1 Claude Opus 4.6 bf16 63.3%
  2. 2 Claude Sonnet 4.6 bf16 55.7%
  3. 3 Qwen3.6 27B fp8 49.3%
  4. 4 gpt-oss-120b mxfp4 31.7%
  5. 5 Nemotron 3 Super nvfp4 30.8%

Source: Official RH SWE-bench v1 leaderboard

#ModelScoreSettingDateSource
1 Claude Opus 4.6 63.3% bf16, Claude Code Feb 2026 Official leaderboard
2 Claude Sonnet 4.6 55.7% bf16, Claude Code Feb 2026 Official leaderboard
3 Qwen3.6 27B 49.3% fp8, Claude Code Apr 2026 Official leaderboard
4 gpt-oss-120b 31.7% mxfp4, Claude Code Aug 2025 Official leaderboard
5 Nemotron 3 Super 30.8% nvfp4, OpenCode Mar 2026 Official leaderboard

The best score, 63.3%, costs $1.22 a task

Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

OpenAI Anthropic Qwen Cost frontier One model at each effort level
0% 20% 40% 60% 80% $0.01 $0.10 $1.00 $10 Cost per task, log scale Claude Opus 4.6, bf16 Claude Sonnet 4.6, bf16 Qwen3.6 27B, fp8 gpt-oss-120b, mxfp4 0% 20% 40% 60% 80% $0.01 $0.10 $1.00 $10 Cost per task, log scale Claude Opus 4.6, bf16 Claude Sonnet 4.6, bf16

Source: Official RH SWE-bench v1 leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 4.6 bf16 Claude Code 63.3% $1.22 7,908
Claude Sonnet 4.6 bf16 Claude Code 55.7% $0.98 13,831
Qwen3.6 27B fp8 Claude Code 49.3% $0.21 13,344
Qwen3.6 27B fp8 Pi 46.8% $0.12 10,136
Qwen3.6 27B fp8 OpenCode 44% $0.12 7,987
gpt-oss-120b mxfp4 Claude Code 31.7% $0.04 17,368
Nemotron 3 Super nvfp4 OpenCode 30.8% $0.10 15,983
gpt-oss-120b mxfp4 OpenCode 29.4% $0.01 6,380
gpt-oss-120b mxfp4 Pi 23% $0.01 5,475
Nemotron 3 Super nvfp4 Claude Code 22.4% $0.11 15,434
Nemotron 3 Super nvfp4 Pi 21.6% $0.09 18,260

Number one over time

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

Anthropic
Own leaderboard Replaced by SWE-bench, Oct 2026

Source: Official RH SWE-bench v1 leaderboard

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
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.6 gpt-oss-120b 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 4.6 gpt-oss-120b

Source: Official RH SWE-bench v1 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

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

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (5 entries).
  • Pass Models scored 5 models we track.
  • Note Newest score The newest came 7 months ago, in April 2026, for Qwen3.6 27B. That’s expected for a test replaced by RH SWE-bench.
  • Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is Claude Opus 4.6 (#39).
  • Pass Headroom The best is 63.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on huggingface.co: 11 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