Doubao Work ranked first at 85.33 in SuperCLUE’s reported October 2026 benchmark. The test focused on finished Word, PowerPoint, Excel and chart deliverables across document processing, materials drafting and data analysis.[1] Doubao led document processing; Kimi Work led materials drafting and data analysis, accord...
Published byEdited with GPT-6 LunaImages generated with GPT Image 2
Research answer

Create a landscape editorial hero image for this Studio Global article: What did SuperCLUE’s October 2026 SuperCLUE-Office end-to-end office-agent benchmark, reported on November 5, reveal about the overall ranki. Article summary: SuperCLUE’s October 2026 office-agent benchmark put Doubao Work first at 85.33, with only 2.65 points separating the top four but a 32.2-point spread across all eight entrants.[7] The available excerpts support that over. Topic tags: general, general web. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fake numbers, clic
SuperCLUE’s reported October 2026 office-agent benchmark placed Doubao Work first with a score of 85.33. The top four were tightly grouped, but the spread across all eight entrants was much wider. The evaluation focused on practical office-file delivery—not just conversational answers—so its results are most useful as a comparison within this test, not as a universal ranking of office agents.1
21
| Rank | Office agent | Overall score |
|---|---|---|
| 1 | Doubao Work | 85.33 |
| 2 | Kimi Work | 84.57 |
| 3 | WorkBuddy | 83.99 |
| 4 | Baidu Dazi | 82.68 |
| 5 | Qianwen Office | 78.38 |
| 6–7 | SenseTime Xiaohuanxiong | Not stated in the available excerpt |
| 6–7 | iFlytek Spark Work | Not stated in the available excerpt |
| 8 | Nano Work | Not stated; the reported 32.2-point top-to-bottom gap implies a score of about 53.1 |
The available report names all eight products, but does not provide the exact scores or ordering of SenseTime Xiaohuanxiong and iFlytek Spark Work. The approximate Nano Work figure is calculated from the reported maximum gap and Doubao Work’s score; it is not a directly quoted score.1
21
The top-four gap is easy to quantify: Doubao led Kimi by 0.76 points, WorkBuddy by 1.34 and Baidu Dazi by 2.65. Across the full field, the reported high-to-low gap was 32.2 points.21
The benchmark covered three task dimensions: document processing, materials drafting and data analysis. It comprised 13 secondary tasks and assessed the completion quality of office deliverables such as Word documents, PowerPoint presentations, Excel spreadsheets and charts.1
SuperCLUE also described five capability areas: file and environment operations; planning and execution across multiple files; following constraints; content quality and professionalism; and safety and permission boundaries.1
The reported results did not identify one agent as the leader in every category. Doubao Work led document processing, while Kimi Work led materials drafting and data analysis.21 The available excerpts do not give the category scores or a full score distribution, so the size of those leads—and how other products compared within each category—cannot be assessed here.
For everyday office work, that distinction matters: an overall score may not tell you which tool best fits a particular workflow. When choosing an agent, compare the files it actually produces on tasks like yours, and check how much review or correction they need. That is a practical way to use this benchmark without assuming its ranking predicts every team’s results.
The results were reported as SuperCLUE’s benchmark findings and repeated in coverage dated November 5, 2026; the available sources do not provide an independent re-test.1
21 They also do not give enough detail here to evaluate category weightings, repeat-run variation or confidence intervals. That makes it especially important not to overread small score differences among the top four.
The narrow conclusion is that Doubao Work led this reported evaluation, with Kimi Work, WorkBuddy and Baidu Dazi close behind. The 32.2-point overall spread suggests a substantially wider separation between the top and bottom of this particular field, but it does not establish that one agent is best for every office workflow.21
Studio Global AI
This page includes a source-backed answer you can continue inside Studio Global.
Doubao Work ranked first at 85.33 in SuperCLUE’s reported October 2026 benchmark.
Doubao Work ranked first at 85.33 in SuperCLUE’s reported October 2026 benchmark. The test focused on finished Word, PowerPoint, Excel and chart deliverables across document processing, materials drafting and data analysis.[1]
Doubao led document processing; Kimi Work led materials drafting and data analysis, according to the reported results.
Doubao Work ranked first at 85.33 in SuperCLUE’s reported October 2026 benchmark. The test focused on finished Word, PowerPoint, Excel and chart deliverables across document processing, materials drafting and data analysis.[1] Doubao led document processing; Kimi Work led materials drafting and data analysis, accord...
Published byEdited with GPT-6 LunaImages generated with GPT Image 2
Research answer

Create a landscape editorial hero image for this Studio Global article: What did SuperCLUE’s October 2026 SuperCLUE-Office end-to-end office-agent benchmark, reported on November 5, reveal about the overall ranki. Article summary: SuperCLUE’s October 2026 office-agent benchmark put Doubao Work first at 85.33, with only 2.65 points separating the top four but a 32.2-point spread across all eight entrants.[7] The available excerpts support that over. Topic tags: general, general web. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fake numbers, clic
SuperCLUE’s reported October 2026 office-agent benchmark placed Doubao Work first with a score of 85.33. The top four were tightly grouped, but the spread across all eight entrants was much wider. The evaluation focused on practical office-file delivery—not just conversational answers—so its results are most useful as a comparison within this test, not as a universal ranking of office agents.1
21
| Rank | Office agent | Overall score |
|---|---|---|
| 1 | Doubao Work | 85.33 |
| 2 | Kimi Work | 84.57 |
| 3 | WorkBuddy | 83.99 |
| 4 | Baidu Dazi | 82.68 |
| 5 | Qianwen Office | 78.38 |
| 6–7 | SenseTime Xiaohuanxiong | Not stated in the available excerpt |
| 6–7 | iFlytek Spark Work | Not stated in the available excerpt |
| 8 | Nano Work | Not stated; the reported 32.2-point top-to-bottom gap implies a score of about 53.1 |
The available report names all eight products, but does not provide the exact scores or ordering of SenseTime Xiaohuanxiong and iFlytek Spark Work. The approximate Nano Work figure is calculated from the reported maximum gap and Doubao Work’s score; it is not a directly quoted score.1
21
The top-four gap is easy to quantify: Doubao led Kimi by 0.76 points, WorkBuddy by 1.34 and Baidu Dazi by 2.65. Across the full field, the reported high-to-low gap was 32.2 points.21
The benchmark covered three task dimensions: document processing, materials drafting and data analysis. It comprised 13 secondary tasks and assessed the completion quality of office deliverables such as Word documents, PowerPoint presentations, Excel spreadsheets and charts.1
SuperCLUE also described five capability areas: file and environment operations; planning and execution across multiple files; following constraints; content quality and professionalism; and safety and permission boundaries.1
The reported results did not identify one agent as the leader in every category. Doubao Work led document processing, while Kimi Work led materials drafting and data analysis.21 The available excerpts do not give the category scores or a full score distribution, so the size of those leads—and how other products compared within each category—cannot be assessed here.
For everyday office work, that distinction matters: an overall score may not tell you which tool best fits a particular workflow. When choosing an agent, compare the files it actually produces on tasks like yours, and check how much review or correction they need. That is a practical way to use this benchmark without assuming its ranking predicts every team’s results.
The results were reported as SuperCLUE’s benchmark findings and repeated in coverage dated November 5, 2026; the available sources do not provide an independent re-test.1
21 They also do not give enough detail here to evaluate category weightings, repeat-run variation or confidence intervals. That makes it especially important not to overread small score differences among the top four.
The narrow conclusion is that Doubao Work led this reported evaluation, with Kimi Work, WorkBuddy and Baidu Dazi close behind. The 32.2-point overall spread suggests a substantially wider separation between the top and bottom of this particular field, but it does not establish that one agent is best for every office workflow.21
Studio Global AI
This page includes a source-backed answer you can continue inside Studio Global.
Doubao Work ranked first at 85.33 in SuperCLUE’s reported October 2026 benchmark.
Doubao Work ranked first at 85.33 in SuperCLUE’s reported October 2026 benchmark. The test focused on finished Word, PowerPoint, Excel and chart deliverables across document processing, materials drafting and data analysis.[1]
Doubao led document processing; Kimi Work led materials drafting and data analysis, according to the reported results.