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

LMArena Text (style-controlled, coding)

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

ModelScoreDateKind
Claude 3.5 Sonnet1313.77409704962662025-04-16snapshot
Llama 3.1 405B Instruct Bf161258.9877215756622025-01-24snapshot
GPT-4O1264.883918251442025-02-17snapshot
Llama 3.1 405B Instruct Fp81255.59699484074832024-12-22snapshot
Grok 21254.30201682961162024-12-18snapshot
Gemini 1.5 Pro Api 05141242.15991121604472024-10-07snapshot
Mistral Large 24071242.98109144025782024-10-07snapshot
GPT-4O Mini1242.69948645804562024-09-27snapshot
GPT-4 Turbo1239.56612129571752024-09-15snapshot
Gemini Advanced 05141238.05678895542592024-09-15snapshot
DeepSeek V2.51255.37656769590332024-12-18snapshot
O1 Preview1311.07779487873272025-04-16snapshot
O1 Mini1291.77047990685382025-03-25snapshot
Gemini 1.5 Pro 0021254.67734200085212024-12-10snapshot
YI Lightning1261.95196243418152025-02-03snapshot
GLM 4 Plus1247.8108012708492024-10-23snapshot
Athene V2 Chat1259.83331329420622025-01-28snapshot
Claude 3.5 Haiku1261.83372271245842025-02-05snapshot
DeepSeek V2.5 12101262.17106712465422025-02-11snapshot
Qwen2 5 Plus 11271258.4025232756532025-01-05snapshot
O11404.81180259523132025-06-01snapshot
DeepSeek V31282.22835579078562025-03-11snapshot
Step 2 16k Exp 2024121259.69860600683392025-01-22snapshot
DeepSeek R11431.93822992712632025-07-07snapshot
Qwen2 51295.91506112576422025-03-26snapshot
Gemini 2.0 Flash 0011279.34693986787992025-02-27snapshot
O3 Mini1308.7581588210542025-04-16snapshot
Qwen Plus 01251273.51820919128612025-02-21snapshot
O3 Mini1389.8585538210592025-05-22snapshot
Claude 3.7 Sonnet1322.200994899122025-04-23snapshot