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

LMArena Text (style-controlled, overall)

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

ModelScoreDateKind
ChatGPT 4O Latest1438.5818799843572025-11-20snapshot
GPT-4O1263.3573505512792025-02-21snapshot
Claude 3.5 Sonnet1285.59087767014062025-04-16snapshot
Gemini Advanced 05141254.3585927155292025-02-05snapshot
Grok 21249.18633846678452024-12-22snapshot
Llama 3.1 405B Instruct1250.04230517455972024-09-04snapshot
GPT-4 Turbo1241.01558149947142024-10-23snapshot
Claude 3 Opus1237.8997881371182024-09-27snapshot
Gemini 1.5 Pro Api 05141237.97005937919042024-10-07snapshot
GPT-4 1106 Preview1234.47885434572192024-09-15snapshot
GPT-4O Mini1230.45906038119032024-09-15snapshot
Llama 3.1 405B Instruct Bf161253.73412452083812025-02-03snapshot
Llama 3.1 405B Instruct Fp81252.84304434049252025-01-28snapshot
O1 Preview1376.40144821854122025-06-11snapshot
O1 Mini1256.28520923366982025-02-06snapshot
Gemini 1.5 Pro 0021268.61299316184662025-03-03snapshot
YI Lightning1246.1242240718582024-12-22snapshot
O11398.97084162394032025-07-07snapshot
DeepSeek V31276.4520960440642025-03-17snapshot
Step 2 16k Exp 2024121252.92065113999842025-01-22snapshot
DeepSeek R11388.12893177712322025-06-11snapshot
Qwen2 51285.68500647127532025-04-16snapshot
Gemini 2.0 Flash 0011283.69335720559322025-03-26snapshot
O3 Mini1266.98901890576282025-02-27snapshot
Gemini 2.0 Flash Lite Preview 02.051266.30584209303132025-02-27snapshot
Qwen Plus 01251265.48072277671332025-02-17snapshot
O3 Mini1286.88012557322052025-04-16snapshot
Claude 3.7 Sonnet1297.5779724067782025-05-11snapshot
GPT-4.5 Preview1442.4612280513062025-12-19snapshot
Grok 3 Preview 02.241409.2012070301062025-08-04snapshot