VECTOR WIREAI INTELLIGENCE
NVDA$1,847+3.2%MSFT$512+1.1%GOOGL$199-0.4%META$728+2.7%AMD$184-1.2%TSM$212+0.6%PLTR$98+4.1%AI IDX4,821+1.9%
PKT
SEEDRefresh Models Deals Regulatory Sources

livecodebench_hard

19 models tracked

LiveCodeBench, Hard difficulty subset — Difficulty-tier subset, never merge with overall/easy/medium scores.

#ModelVendorBest scoreRunsLast seen
1o3OpenAI6612026-05-03
2DeepSeek-R1DeepSeek63.612026-05-03
3o4-miniOpenAI62.932026-05-03
4Gemini 2.5 Pro Preview 06.05 (32k think)Google6212026-05-30
5Gemini 2.5 ProGoogle59.422026-05-03
6OpenReasoning-Nemotron-32BNVIDIA57.512026-05-03
7EXAONE 4.0 32BLG AI56.312026-05-03
8Qwen3 235B A22BAlibaba5412026-05-03
9OpenCodeReasoning-Nemotron-1.1-32BNVIDIA52.812026-05-03
10Grok 3 Mini ReasoningSpaceXAI51.112026-05-03
11Gemini 2.5 FlashGoogle48.332026-06-20
12o3-miniOpenAI47.432026-06-18
13Claude 4 OpusAnthropic33.122026-05-29
14Claude Sonnet 4Anthropic32.622026-05-15
15DeepSeek-V3DeepSeek14.212026-05-03
16GPT-4oOpenAI4.512026-05-03
17GPT-4o miniOpenAI4.312026-05-03
18GPT-4 TurboOpenAI412026-05-03
19Claude 3 HaikuAnthropic1.912026-05-03

Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.