evals.report
BenchmarksLabsCompareRun guidesIn the wild
BenchmarksCoding

LiveCodeBench Pro

A live competitive-programming benchmark that rates LLMs with a Codeforces-style Elo on fresh contest problems.

CodingCodeforces EloHigher is better
ScoresAboutRun this benchmark

What is LiveCodeBench Pro?

A live competitive-programming benchmark that rates LLMs with a Codeforces-style Elo on fresh contest problems. evals.report tracks reported LiveCodeBench Pro scores with the model, source, status, date, and run caveats attached — official leaderboard scores, vendor-reported launches, and clearly labeled community runs.

Top reported LiveCodeBench Pro score: Gemini 3 Deep Think — 3298 (Codeforces Elo).

ModelLabScore↓Source modelStatusDate
Gemini 3 Deep ThinkGoogle DeepMind3298Gemini 3 Deep ThinkOfficialDec 4, 2025Details
DeepSeek V4 ProOpenDeepSeek3206DeepSeek V4 Pro MaxUnverifiedApr 24, 2026Details
Gemini 3.1 Pro PreviewGoogle DeepMind2887Gemini 3.1 Pro PreviewOfficialFeb 19, 2026Details
Gemini 3 ProGoogle DeepMind2439Gemini 3 Pro PreviewOfficialNov 18, 2025Details
GPT-5.2OpenAI2393GPT-5.2-high (2025-12-11)OfficialDec 11, 2025Details
Gemini 3 FlashGoogle DeepMind2316Gemini 3 Flash PreviewOfficialDec 17, 2025Details
GPT-5.1OpenAI2269GPT-5.1-high (2025-11-13)OfficialNov 12, 2025Details
GPT-5OpenAI2176GPT-5-high (2025-08-07)OfficialAug 7, 2025Details
o4-miniOpenAI2092o4-mini-high (2025-04-16)OfficialApr 16, 2025Details
Gemini 2.5 ProGoogle DeepMind1769Gemini 2.5 ProOfficialMar 25, 2025Details
Qwen3 235B A22B Instruct 2507OpenAlibaba / Qwen1673Qwen3-235B-A22B-Thinking-2507OfficialJul 21, 2025Details
Claude Sonnet 4.5Anthropic1412Claude 4.5 Sonnet ThinkingOfficialSep 29, 2025Details
GPT-OSS-120BOpenOpenAI1299GPT OSS 120BOfficialAug 5, 2025Details
Gemini 2.5 FlashGoogle DeepMind1288Gemini 2.5 FlashOfficialApr 17, 2025Details
DeepSeek R1OpenDeepSeek1284DeepSeek R1 (2025-05-28)OfficialJan 20, 2025Details
Qwen3 MaxAlibaba / Qwen1226Qwen3-MaxOfficialSep 5, 2025Details
DeepSeek V3 0324OpenDeepSeek1124DeepSeek V3 (2025-03-24)OfficialMar 24, 2025Details
o3OpenAI1010o3-high (2025-04-16)OfficialApr 16, 2025Details
GPT-4.1OpenAI606GPT-4.1OfficialApr 14, 2025Details
Claude 3.5 SonnetAnthropic572Claude 3.5 SonnetOfficialJun 20, 2024Details
Llama 4 MaverickOpenMeta528Llama 4 MaverickOfficialApr 5, 2025Details
GPT-4oOpenAI210GPT-4o (2024-11-20)OfficialMay 13, 2024Details

Each row reports the model’s Codeforces Elo on LiveCodeBench Pro. Click a row for the full run context.