LMArena Agent Arena Steering Burden
Agents
Signed percentage-point Agent Arena steering_burden diagnostic, displayed by the source as Steerability. Measures response to user corrections. Kept separate from the earlier steerability source signal because its key and observation population changed on the September 30 snapshot.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 4 Argon | 15.9 |
| GPT-6 Sol | 11.9 |
| Claude Fable 5 | 9.0 |
| Claude Fable 5.1 | 9.0 |
| GPT-6 Astra | 8.2 |
| Opus 4.8 | 7.5 |
| Claude Opus 5 | 7.0 |
| GPT-5.6 Sol | 6.2 |
| Muse Spark 1.3 | 5.5 |
| GPT-5.4 | 5.4 |
| Sonnet 5 | 5.4 |
| GPT-5.5 | 3.9 |
| Gemini 3.8 Flash | 3.3 |
| Qwen3.8-Max | 3.1 |
| GPT-5.6 Terra | 2.5 |
| Grok 4.5 | 2.1 |
| GLM-5.2 | 1.8 |
| Kimi K3 | 1.4 |
| DeepSeek V4.1 Flash | 1.2 |
| Grok 4.6 | 1.0 |
| Hy4 Preview | 0.9 |
| GLM-5.3 | 0.9 |
| Grok 4.7 | 0.4 |
| Hy3 | 0.4 |
| DeepSeek V4 Pro 0813 | 0.3 |
| GPT-5.6 Luna | 0.1 |
| GLM-5.3 Flash | -0.2 |
| Qwen3.8 27B | -0.4 |
| Qwen3.8-Flash-Next | -0.8 |
| Gemini 3.7 Flash | -1.0 |
| Gemini 3.1 Pro Preview | -4.5 |
| Muse Spark 1.2 | -4.9 |
| Muse Spark 1.1 | -6.2 |
| Qwen3.7-Max | -6.4 |
| MiniMax M3 | -6.9 |
| MiMo-V2.5-Pro | -7.4 |
| Gemini 3.6 Flash | -7.5 |
| Qwen3.7-Plus | -8.4 |
| Solar Pro 4 | -9.1 |
| Inkling | -11.9 |