Benchmark
CoWorkBench
Scores are percentages. Higher is better.
- Models measured
- 12
- Scores recorded
- 23
- Sources
- 2
- IQ weight
- 1.27
Score distribution
- 69.3%
- Mean
- 70.7%
- Median
- 6.3%
- Std dev
- 65.1%
- 10th percentile
- 74.8%
- 90th percentile
- 45.1%
- Min
This benchmark has a weight of 1.27 for chat models, measured over 12 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5Qwen内部 CoWorkBench· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 75.9% |
| 2 | Qwen3.8 Omni FlashQwen 内部长程办公与生产力任务· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 75.3% |
| 3 | Qwen3.8 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 74.8% |
| 4 | Qwen3.8 MaxQwen内部 CoWorkBench· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 74.8% |
| 5 | Qwen 3.8 Flash Next· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 73.9% |
| 6 | Qwen 3.8 Flash NextQwen3.8-Flash-Next release/model card · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 73.9% |
| 7 | Qwen3.8 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 73.9% |
| 8 | Qwen3.8 FlashQwen 内部长程办公与生产力任务· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 73.9% |
| 9 | Claude Opus 4.8Qwen内部 CoWorkBench· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 72.3% |
| 10 | GPT-5.6 SolQwen内部 CoWorkBench· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 71.5% |
| 11 | Qwen 3.8 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 70.7% |
| 12 | Qwen 3.8 27BQwen 内部长程办公与生产力任务· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 70.7% |
| 13 | Qwen 3.8 27BQwen3.8-27B model card · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 70.7% |
| 14 | Qwen3.7 MaxQwen3.7 release cross-harness chart · Hermes· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 68.3% |
| 15 | Claude Opus 4.6Qwen 内部长程办公与生产力任务 · max effort· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 68.2% |
| 16 | Qwen3.7 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 67.2% |
| 17 | Qwen3.7 MaxQwen3.7-Max release · OpenClaw · release table and cross-harness figure· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 67.2% |
| 18 | Qwen3.7 MaxQwen3.7 release cross-harness chart · Claude Code· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 66% |
| 19 | Qwen3.7 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 65.1% |
| 20 | Qwen3.7 PlusQwen 内部长程办公与生产力任务· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 65.1% |
| 21 | Qwen3.7 PlusQwen3.7-Plus release · release table · harness not separately disclosed· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 65.1% |
| 22 | Qwen3.7 MaxQwen内部 CoWorkBench· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 64.6% |
| 23 | DeepSeek V4 Flash 0731Qwen 内部长程办公与生产力任务· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 45.1% |
Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats