Benchmark Almanac

Factory Code Review Benchmark

GPT-5.2 60.5% leads since April 2026
Not in model cards

Models review real pull requests. Factory measures precision and recall against reference issues and reports mean F1.

The launch study includes 13 models and 50 pull requests.

Made by
Factory
Released
April 2026
Kind
Fixing real repositories
Level
Expert
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Code → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
Strongest model tested
GPT-5.5 #21 on the capability index
Last model added
April 2026
Scores refreshed
November 20th, 2026
13
Models scored on its own leaderboard

GPT-5.2 leads with 60.5% 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 GPT-5.2 60.5%
  2. 2 Claude Opus 4.6 59.8%
  3. 3 Claude Sonnet 4.6 57.4%
  4. 4 Claude Opus 4.7 55.9%
  5. 5 GLM-5.1 55.8%
  6. 6 GPT-5.3 Codex 55.7%
  7. 7 Gemini 3.1 Pro 52.1%
  8. 8 Kimi K2.5 51.9%
  9. 9 GPT-5.4 Mini 51.5%
  10. 10 Gemini 3 Flash 49.5%
  11. 11 GPT-5.5 47.9%
  12. 12 GPT-5.4 47.5%
  13. 13 MiniMax-M2.7 45.6%

Source: Official Factory Code Review Benchmark leaderboard

#ModelScoreSettingDateSource
1 GPT-5.2 60.5% Dec 2025 Official leaderboard
2 Claude Opus 4.6 59.8% Feb 2026 Official leaderboard
3 Claude Sonnet 4.6 57.4% Feb 2026 Official leaderboard
4 Claude Opus 4.7 55.9% Apr 2026 Official leaderboard
5 GLM-5.1 55.8% Apr 2026 Official leaderboard
6 GPT-5.3 Codex 55.7% Feb 2026 Official leaderboard
7 Gemini 3.1 Pro 52.1% Feb 2026 Official leaderboard
8 Kimi K2.5 51.9% Jan 2026 Official leaderboard
9 GPT-5.4 Mini 51.5% Mar 2026 Official leaderboard
10 Gemini 3 Flash 49.5% Dec 2025 Official leaderboard
11 GPT-5.5 47.9% Apr 2026 Official leaderboard
12 GPT-5.4 47.5% Mar 2026 Official leaderboard
13 MiniMax-M2.7 45.6% Mar 2026 Official leaderboard

The best score, 60.5%, costs $1.25 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 Google DeepMind Moonshot AI Anthropic MiniMax Z.ai Cost frontier One model at each effort level
0% 20% 40% 60% 80% $0.10 $0.20 $0.50 $1.00 $2.00 $5.00 $10 Cost per task, log scale GPT-5.2 GLM-5.1 Kimi K2.5 MiniMax-M2.7 0% 20% 40% 60% 80% $0.10 $1.00 $10 Cost per task, log scale GPT-5.2 GLM-5.1

Source: Official Factory Code Review Benchmark leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
GPT-5.2 60.5% $1.25
Claude Opus 4.6 59.8% $3.11
Claude Sonnet 4.6 57.4% $1.15
Claude Opus 4.7 55.9% $4.18
GLM-5.1 55.8% $1.06
GPT-5.3 Codex 55.7% $1.69
Gemini 3.1 Pro 52.1% $2.04
Kimi K2.5 51.9% $0.41
GPT-5.4 Mini 51.5% $0.68
Gemini 3 Flash 49.5% $0.34
GPT-5.5 47.9% $5.63
GPT-5.4 47.5% $2.01
MiniMax-M2.7 45.6% $0.15

Number one over time

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

OpenAI
Own leaderboard Not tested since Apr 2026
GPT-5.2 60.5%

Source: Official Factory Code Review Benchmark 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
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) GPT-5.2 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) GPT-5.2

Source: Official Factory Code Review Benchmark 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

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 7 months ago, in April 2026, for GPT-5.5. 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 (13 entries).
  • Pass Models scored 13 models we track.
  • Warning Newest score The newest came 7 months ago, in April 2026, for GPT-5.5.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.5 (#21).
  • Pass Headroom The best is 60.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on factory.com: 13 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