Benchmark

LiveCodeBench v5

LiveCodeBench is a holistic and contamination-free evaluation benchmark for large language models for code. It continuously collects new problems from…

Modality
text
Categories
reasoning, general
Openness
unknown
Source
llm_stats
Reported scores
9

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Gemini 2.0 Flash-LiteGoogle0.2892025-02-05evidence
Gemini 2.5 FlashGoogle0.6392025-05-20evidence
Gemini 2.5 ProGoogle0.7562025-05-20evidence
Gemma 3n E2B InstructedGoogle0.1862025-06-26evidence
Gemma 3n E2B Instructed LiteRT (Preview)Google0.1862025-05-20evidence
Gemma 3n E4B InstructedGoogle0.2572025-06-26evidence
Gemma 3n E4B Instructed LiteRT PreviewGoogle0.2572025-05-20evidence
MiniCPM-SALAOpenBMB0.60482026-02-11evidence
Qwen3 VL 235B A22B InstructQwen0.6142025-09-22evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.