Benchmark Almanac

SWE-CoWork

GPT-6 Astra 7.92% leads since November 2026
Not in model cards

Coding agents join a simulated engineering team on 101 real repositories and must gather scattered requirements from simulated colleagues and workplace records before implementing them. Hidden tests score the share of new-behavior tests passed and whether the whole task passes.

Tasks span 11 languages and 12 domains, with 2,712 requirements, 82% of them held only by colleagues, and an 8-hour budget per task. The board ranks by F2P test pass rate. Full-task pass, which needs every F2P and P2P test to pass in one run, stayed under 8% for all 11 models tested.

Made by
TokenWave AI
Released
November 2026
Kind
Fixing real repositories
Level
Expert
Scored
Checked automatically
Answers
Code
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Code → Code, Actions
Interaction
Multi-agent
Environment
Terminal, APIs and tools
Refresh
One fixed set
Run by
Anyone
In views
Agents
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
September 2026
Scores refreshed
November 18th, 2026
11
Entries on its own leaderboard, 10 models we track

GPT-6 Astra leads with 7.92% among models released by September 2026 GPT-6 Astra leads with 7.92% among models released by September 2026

Best score per entry from the benchmark's own leaderboard, read on November 18th, 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. This view shows only models we track, each at its best.

The newest model with a score here came out in September 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-6 Astra CoWork; 101 tasks 7.92%
  2. 2 Claude Opus 5.5 CoWork; 101 tasks 6.93%
  3. 3 Seed Evolving CoWork; 101 tasks 0.99%
  4. 4 Kimi K3 CoWork; 101 tasks 0.99%
  5. 5 GLM-5.3 CoWork; 101 tasks 0.99%
  6. 6 DeepSeek-V4-Pro CoWork; 101 tasks 0%
  7. 7 GLM-5.2 CoWork; 101 tasks 0%
  8. 8 Seed2.1 Pro CoWork; 101 tasks 0%
  9. 9 Qwen 3.8 Max CoWork; 101 tasks 0%
  10. 10 Qwen3.8 Flash CoWork; 101 tasks 0%
  11. 11 DeepSeek V4.1 Flash CoWork; 101 tasks 0%

Source: Official SWE-CoWork leaderboard

#EntryScoreSettingDateSource
1 GPT-6 Astra 7.92% CoWork; 101 tasks Sep 2026 Official leaderboard
2 Claude Opus 5.5 6.93% CoWork; 101 tasks Sep 2026 Official leaderboard
3 Seed Evolving 0.99% CoWork; 101 tasks Official leaderboard
4 Kimi K3 0.99% CoWork; 101 tasks Jul 2026 Official leaderboard
5 GLM-5.3 0.99% CoWork; 101 tasks Aug 2026 Official leaderboard
6 DeepSeek-V4-Pro 0% CoWork; 101 tasks Apr 2026 Official leaderboard
7 GLM-5.2 0% CoWork; 101 tasks Jun 2026 Official leaderboard
8 Seed2.1 Pro 0% CoWork; 101 tasks Jun 2026 Official leaderboard
9 Qwen 3.8 Max 0% CoWork; 101 tasks Aug 2026 Official leaderboard
10 Qwen3.8 Flash 0% CoWork; 101 tasks Aug 2026 Official leaderboard
11 DeepSeek V4.1 Flash 0% CoWork; 101 tasks Sep 2026 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-6 Astra 7.92% CoWork; 101 tasks Sep 2026 Official leaderboard
2 Claude Opus 5.5 6.93% CoWork; 101 tasks Sep 2026 Official leaderboard
3 Kimi K3 0.99% CoWork; 101 tasks Jul 2026 Official leaderboard
4 GLM-5.3 0.99% CoWork; 101 tasks Aug 2026 Official leaderboard
5 DeepSeek-V4-Pro 0% CoWork; 101 tasks Apr 2026 Official leaderboard
6 GLM-5.2 0% CoWork; 101 tasks Jun 2026 Official leaderboard
7 Seed2.1 Pro 0% CoWork; 101 tasks Jun 2026 Official leaderboard
8 Qwen 3.8 Max 0% CoWork; 101 tasks Aug 2026 Official leaderboard
9 Qwen3.8 Flash 0% CoWork; 101 tasks Aug 2026 Official leaderboard
10 DeepSeek V4.1 Flash 0% CoWork; 101 tasks Sep 2026 Official leaderboard

Number one over time

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

OpenAI
Own leaderboard 1 leader
GPT-6 Astra 7.92%

Source: Official SWE-CoWork 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%) GPT-6 Astra Kimi K3 DeepSeek-V4-Pro 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) GPT-6 Astra Kimi K3

Source: Official SWE-CoWork 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

All 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (11 entries).
  • Pass Models scored 10 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from September 2026, for Claude Opus 5.5.
  • Note Frontier models Only 2 of the top 10 have a score. The strongest tested is GPT-6 Astra (#4).
  • Pass Headroom The best is 7.92%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on tokenwave-ai.github.io: 10 rows for models we track and 1 other entry at the last read.
  • Pass Last read November 18th, 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