Agentic benchmark
CoWorkBench leaderboard
Every model the catalog carries a published CoWorkBench value for, ranked by that value.
CategoryAgentic
MeasureProvider-run agent score
TasksLong-horizon professional workflows
DifficultyCross-domain professional work
Published byQwen3.8-Max: A New Bar for Coding and Cowork
Qwen's internal benchmark for long-horizon professional work across computer science, finance, law, medicine, and other productivity domains.
CoWorkBench ranking
4 models with a published CoWorkBench value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Provider-run agent score |
|---|---|---|---|
| 1 |  Qwen3.8-Omni-Flash | Alibaba | 75.3 |
| 2 | Qwen3.8 Max | Alibaba | 74.8 |
| 3 | Qwen3.8-Flash-Next | Alibaba | 73.9 |
| 4 | Qwen3.8-27B | Alibaba | 70.7 |
Evidence key: Observed
Rows are ordered by the value Qwen3.8-Max: A New Bar for Coding and Cowork published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.