LMArena Agent Arena Chat Praise vs Complaint
Agents · 2026-08-14
Signed percentage-point Agent Arena Chat category diagnostic signal comparing positive user feedback against complaints, recomputed on sessions tagged as general agentic chat such as creative writing, learning, everyday research and media generation. 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 | 46.8 |
| Claude Fable 5.1 | 30.9 |
| Gemini 4 Argon | 30.2 |
| Claude Sonnet 5.5 | 27.4 |
| Muse Wonder Full | 26.7 |
| GPT-6.1 Sol | 19.9 |
| GPT-5.6 Sol | 19.6 |
| GPT-6 Astra | 18.4 |
| Claude Fable 5 | 18.1 |
| Claude Opus 5 | 15.3 |
| Gemini 3.8 Flash | 12.5 |
| GPT-6 Sol | 11.3 |
| GLM-5.3 | 7.1 |
| Hy4 Preview | 6.2 |
| GLM-5.2 | 5.5 |
| MiMo-V2.6-Pro | 5.2 |
| Muse Spark 1.3 | 4.0 |
| Gemini 3.7 Flash | 3.5 |
| DeepSeek V4.1 Flash | 3.5 |
| Kimi K3 | 2.8 |
| GPT-5.6 Terra | 2.8 |
| GLM-5.3 Flash | 2.1 |
| Gemini 3.1 Pro Preview | 1.8 |
| DeepSeek V4 Pro 0813 | 1.2 |
| Qwen3.8-Max | -0.2 |
| Grok 4.6 | -0.3 |
| Grok 4.7 | -0.8 |
| MiMo-V2.6-Flash | -1.3 |
| GPT-5.4 | -1.6 |
| Mistral Large 4 Preview | -1.8 |
| GPT-6 Luna | -2.2 |
| GPT-5.5 | -2.9 |
| Muse Wonder | -4.1 |
| Qwen3.8-Flash-Next | -4.7 |
| GPT-5.6 Luna | -5.9 |
| DeepSeek V4.1 Flash 1017 | -6.8 |
| Sonnet 5 | -7.3 |
| Muse Spark 1.1 | -7.8 |
| Qwen3.8 27B | -8.2 |
| Step 5 Preview | -8.6 |