LMArena Agent Arena Work Praise vs Complaint
Agents · 2026-08-14
Signed percentage-point Agent Arena Work category diagnostic signal comparing positive user feedback against complaints, recomputed on sessions tagged as office work such as document creation, professional research, planning and professional writing. Values may be negative relative to the category baseline; reported confidence intervals are not standard errors.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5.5 | 46.3 |
| GPT-6.1 Sol | 33.4 |
| GPT-6 Astra | 32.2 |
| Claude Fable 5.1 | 27.3 |
| Gemini 4 Argon | 27.0 |
| Claude Fable 5 | 16.0 |
| Claude Sonnet 5.5 | 15.2 |
| Muse Wonder Full | 14.2 |
| GPT-6 Sol | 13.5 |
| GPT-5.6 Sol | 13.1 |
| Claude Opus 5 | 10.8 |
| DeepSeek V4.1 Flash 1017 | 9.0 |
| GPT-5.5 | 8.3 |
| Sonnet 5 | 8.1 |
| GLM-5.2 | 6.3 |
| MiMo-V2.6-Pro | 5.9 |
| GLM-5.3 | 4.8 |
| MiMo-V2.6-Flash | 4.7 |
| Hy4 Preview | 4.3 |
| DeepSeek V4.1 Flash | 3.9 |
| Gemini 3.8 Flash | 3.9 |
| Muse Spark 1.3 | 3.5 |
| Grok 4.7 | 3.0 |
| GPT-5.6 Terra | 2.6 |
| Qwen3.8-Max | 2.4 |
| Kimi K3 | 2.1 |
| GPT-6 Luna | 1.0 |
| DeepSeek V4 Pro 0813 | -0.1 |
| Muse Wonder | -0.5 |
| Step 5 Preview | -0.9 |
| GPT-5.4 | -1.1 |
| Grok 4.6 | -2.1 |
| GLM-5.3 Flash | -2.8 |
| Claude Haiku 5.5 | -3.0 |
| Mistral Large 4 Preview | -3.1 |
| Qwen3.8-Flash-Next | -3.4 |
| Gemini 3.1 Pro Preview | -3.6 |
| Gemini 3.7 Flash | -3.9 |
| GPT-5.6 Luna | -4.6 |
| Qwen3.8 27B | -5.3 |