LMArena Agent Arena Code Praise vs Complaint
Agents · 2026-08-14
Signed percentage-point Agent Arena Code category diagnostic signal comparing positive user feedback against complaints, recomputed on sessions tagged as writing or debugging code, automating workflows or analyzing data. 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 | 51.3 |
| GPT-6 Astra | 36.6 |
| GPT-6.1 Sol | 31.9 |
| Claude Fable 5.1 | 27.4 |
| Gemini 4 Argon | 22.1 |
| Claude Sonnet 5.5 | 19.6 |
| Muse Wonder Full | 16.0 |
| Claude Fable 5 | 15.1 |
| GPT-6 Sol | 13.0 |
| GPT-5.5 | 12.3 |
| DeepSeek V4.1 Flash 1017 | 11.2 |
| GPT-5.6 Sol | 10.2 |
| Sonnet 5 | 8.0 |
| Claude Opus 5 | 7.8 |
| GLM-5.2 | 5.8 |
| GLM-5.3 | 4.1 |
| MiMo-V2.6-Flash | 4.1 |
| Grok 4.7 | 3.9 |
| Hy4 Preview | 3.3 |
| DeepSeek V4.1 Flash | 3.3 |
| MiMo-V2.6-Pro | 2.9 |
| Muse Spark 1.3 | 2.8 |
| GPT-6 Luna | 2.5 |
| Gemini 3.8 Flash | 2.4 |
| GPT-5.6 Terra | 2.3 |
| Qwen3.8-Max | 1.8 |
| Claude Haiku 5.5 | 1.5 |
| GPT-5.4 | 0.4 |
| Kimi K3 | 0.3 |
| DeepSeek V4 Pro 0813 | -1.1 |
| Muse Wonder | -1.3 |
| Grok 4.6 | -1.3 |
| Step 5 Preview | -1.8 |
| Qwen3.8-Flash-Next | -2.6 |
| Hy3 | -3.2 |
| Gemini 3.7 Flash | -3.4 |
| GLM-5.3 Flash | -3.4 |
| Mistral Large 4 Preview | -3.9 |
| Qwen3.8 27B | -4.9 |
| GPT-5.6 Luna | -5.4 |