⌕Is ___ SOTA?
benchmarks / lmarena-text-long

LMArena Text (style-controlled, longer query)

Arena score. Status active. Contamination risk low. Maintainer LMArena. Source.

ModelScoreDateKind
O11412.49607747674082026-10-08snapshot
Claude 3.5 Sonnet1396.33845572439992026-10-08snapshot
O1 Preview1382.22672271324682026-10-08snapshot
DeepSeek V31374.8483765224792026-10-08snapshot
Gemini 1.5 Pro 0021292.72788657744292025-02-27snapshot
O1 Mini1282.24918462663442025-02-11snapshot
Gemini 1.5 Pro 0011279.49314056253252025-02-09snapshot
Claude 3.5 Haiku1271.90179457285032025-02-03snapshot
DeepSeek V2.5 12101270.0347378681112025-01-28snapshot
GPT-4O1269.72411949107822025-01-22snapshot
Grok 21268.58258223400232025-01-22snapshot
Step 2 16k Exp 2024121276.7514998052672025-02-05snapshot
DeepSeek R11399.51585013010782026-10-08snapshot
Qwen2 51384.94129810924262026-10-08snapshot
Gemini 2.0 Flash 0011309.97848728640542025-03-17snapshot
O3 Mini1307.21193995163372025-03-03snapshot
Gemini 2.0 Flash Lite Preview 02.051283.34859798454542025-02-17snapshot
Qwen Plus 01251292.61925344836352025-02-21snapshot
O3 Mini1373.1313793275442026-10-08snapshot
Claude 3.7 Sonnet1412.14137994445952026-10-08snapshot
Hunyuan Large1376.43155532509452026-10-08snapshot
GPT-4.5 Preview1443.26876168501262026-10-08snapshot
Grok 3 Preview 02.241428.0335228245012026-10-08snapshot
Claude 3.7 Sonnet 20250219 Thinking1429.67539662527092026-10-08snapshot
Command A 03 20251370.27443950165642026-10-08snapshot
ChatGPT 4O Latest1439.5819868382472026-10-08snapshot
DeepSeek V3 03241393.5956397395172026-10-08snapshot
GPT-4.11426.40717582304112026-10-08snapshot
O31411.32647702386522026-10-08snapshot
GPT-4.1 Mini1389.82067607796262026-10-08snapshot