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 | 51.9 |
| Claude Sonnet 5.5 | 37.3 |
| Claude Fable 5.1 | 32.0 |
| Gemini 4 Argon | 30.2 |
| GPT-6.1 Sol | 25.2 |
| Kimi K3.1 Pro | 23.9 |
| Muse Wonder Full | 21.3 |
| Claude Opus 5 | 18.2 |
| GPT-5.6 Sol | 15.0 |
| GPT-6 Astra | 14.3 |
| Gemini 3.8 Flash | 13.9 |
| Claude Fable 5 | 12.9 |
| Hy4 Preview | 5.7 |
| GLM-5.3 | 5.6 |
| GPT-6 Sol | 5.5 |
| MiMo-V2.6-Pro | 5.3 |
| Muse Spark 1.3 | 4.6 |
| Gemini 3.7 Flash | 3.9 |
| Grok 4.7 | 3.9 |
| GLM-5.2 | 3.6 |
| Kimi K3 | 3.4 |
| DeepSeek V4.1 Flash | 3.1 |
| GPT-5.6 Terra | 2.4 |
| DeepSeek V4.1 Flash 1017 | 0.7 |
| DeepSeek V4 Pro 0813 | 0.6 |
| Gemini 3.1 Pro Preview | -0.6 |
| GLM-5.3 Flash | -0.7 |
| Grok 4.6 | -1.1 |
| Qwen3.8-Max | -1.4 |
| GPT-6 Luna | -2.6 |
| MiMo-V2.6-Flash | -3.3 |
| Muse Wonder | -3.8 |
| Sonnet 5 | -4.5 |
| Qwen3.8-Flash-Next | -4.6 |
| Mistral Large 4 Preview | -5.1 |
| Qwen3.7-Max (source unspecified) | -8.8 |
| Qwen3.8 27B | -9.3 |
| Muse Spark 1.1 | -10.0 |
| MiMo-V2.5-Pro | -10.4 |
| Muse Spark 1.2 | -10.7 |