CoWorkBench
Qwen3.8 Max (0902) 76.1% leads since September 2026 Agent completes long-horizon productivity tasks across computer science, finance, law and medicine. Qwen in-house scoring (method unpublished).
Lab-created in-house eval (Alibaba Qwen). Only Qwen reports scores. Tasks and graders are not public.
- Made by
- Alibaba Qwen
- Kind
- Tasks from real jobs
- Level
- Expert
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- Kept private
- Strongest model tested
- Claude Fable 5 #11 on the capability index
- Last in a launch post
- September 2026
- Last model added
- September 2026
- Scores refreshed
- November 12th, 2026
Claude Fable 5 leads with 75.9% among models released by August 2026
Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.
Qwen3.8 Max (0902) leads with 76.1%, claims included
Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.
The newest model with a score here came out in August 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- Claude Fable 5.2
-
Gemini 4 Carbon - GPT-6 Astra
- Claude Opus 5.5
- and 14 more
- 1 Claude Fable 5 75.9%
- 2 Qwen 3.8 Max 74.8%
- 3 Claude Opus 4.8 72.3%
- 4 GPT-5.6 Sol max 71.5%
- 5 Qwen3.7-Max 64.6%
Source: Official CoWorkBench leaderboard
- 1 Qwen3.8 Max (0902) 76.1%
- 2 Claude Fable 5 75.9%
- 3 Qwen3.8-Omni-Flash 75.3%
- 4 Qwen 3.8 Max 74.8%
- 5 Qwen3.8 Flash 73.9%
- 6 Claude Opus 4.8 72.3%
- 7 GPT-5.6 Sol max 71.5%
- 8 Qwen 3.8 27B 70.7%
- 9 Qwen3.7-Max 67.2%
- 10 Qwen3.7-Plus 65.1%
Sources: Labs’ model cards and official CoWorkBench leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Fable 5 | 75.9% | Jun 2026 | Official leaderboard | |
| 2 | Qwen 3.8 Max | 74.8% | Aug 2026 | Official leaderboard | |
| 3 | Claude Opus 4.8 | 72.3% | May 2026 | Official leaderboard | |
| 4 | GPT-5.6 Sol | 71.5% | max | Jul 2026 | Official leaderboard |
| 5 | Qwen3.7-Max | 64.6% | May 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Qwen3.8 Max (0902) | 76.1% | Sep 2026 | Lab model card | |
| 2 | Claude Fable 5 | 75.9% | Jun 2026 | Official leaderboard | |
| 3 | Qwen3.8-Omni-Flash | 75.3% | Sep 2026 | Lab model card | |
| 4 | Qwen 3.8 Max | 74.8% | Aug 2026 | Official leaderboard | |
| 5 | Qwen3.8 Flash | 73.9% | Aug 2026 | Lab model card | |
| 6 | Claude Opus 4.8 | 72.3% | May 2026 | Official leaderboard | |
| 7 | GPT-5.6 Sol | 71.5% | max | Jul 2026 | Official leaderboard |
| 8 | Qwen 3.8 27B | 70.7% | Aug 2026 | Lab model card | |
| 9 | Qwen3.7-Max | 67.2% | May 2026 | Lab model card | |
| 10 | Qwen3.7-Plus | 65.1% | Jun 2026 | Lab model card |
Labs report scores for 7 models in their cards. Switch the scores above to rank them together.
3 models have held number one on the official leaderboard since 2023 4 models have held number one with claims included since 2023
Lab claims count from the day the card came out, independent runs from the model’s release. We draw one row each for the benchmark’s own leaderboard and what labs claim in their cards.
Sources: Official CoWorkBench leaderboard and labs’ model cards
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Sources: Official CoWorkBench leaderboard and labs’ model cards
One lab put it in a launch post in the last six months
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| Sep 18, 2026 | Qwen3.8-Omni-Flash | 75.3 | CoWorkBench Long-horizon office work |
| Sep 2, 2026 | Qwen3.8 Max (0902) | 76.1 | Long-horizon office work, CoWorkBench |
| Aug 26, 2026 | Qwen3.8 Flash | 73.9 | Long-horizon office work CoWorkBench |
| Aug 14, 2026 | Qwen 3.8 27B | 70.7 | Long-horizon office work CoWorkBench |
| Aug 3, 2026 | Qwen 3.8 Max | 74.8 | CoWorkBench |
| Jun 1, 2026 | Qwen3.7-Plus | 65.1 | CoWorkBench |
| May 20, 2026 | Qwen3.7-Max | 67.2 | CoWorkBench |
One system card reports it
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Sep 22, 2026 | Qwen3.8-Omni-Flash technical report | 75.3 | CoWorkBench, in-house long-horizon office tasks (Table 2) | Capabilities |
2 lab claims next to the leaderboard
Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. 1 of them match, which usually means one side copied the other. Every benchmark.
| Model | Lab says | Board says | Gap |
|---|---|---|---|
| Qwen3.7-Max | 67.2% CoWorkBench | 64.6% Leaderboard | +2.6 pts |
| Qwen 3.8 Max | 74.8% CoWorkBench | 74.8% Leaderboard | Same |
More in Economic work
- JobBench In use 130 tasks from 35 occupations workers most want delegated. Agent works through messy reference files, graded by fact-anchored binary rubrics. GLM-5.3 61.4%
- GDPval In use Real work tasks from 44 occupations with reference files. Experts blindly compare model deliverables with expert work. Metric is win-or-tie rate.
Seed2.1 Pro 87.9% - GDPval-AA v2.1 In use Agents with shell and web access complete 220 GDPval gold tasks from 44 occupations. Blind pairwise comparisons give Elo ratings. Claude Opus 5.5 1,846
- AA-Briefcase v1.1 In use Version 1.1 of Artificial Analysis's multi-hour office work evaluation, with Elo ratings fitted by Crowd-BT. Claude Opus 5.5 1,822
- OfficeQA In use Grounded reasoning over a large corpus of U.S. Treasury Bulletin documents: locate tables and compute precise numeric answers. Claude Opus 4.7 86.3%
- SpreadsheetBench In use Real spreadsheet manipulation questions from Excel forums. Output workbook is checked cell-by-cell against reference solutions.
MiniMax-M3 89.3%