Compare AI models

Specifications
Parameters
236B9B
Context window
128k128k
Benchmarks
LiveCodeBench
41.8%28.5%
HumanEval
89%82.3%
Overview
CompanyDeepSeekZ.ai
Release dateSep 5 2024Jul 5 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V2.5 leads CodeGeeX4-ALL-9B on 2 of the 2 benchmarks they both report (LiveCodeBench, HumanEval). CodeGeeX4-ALL-9B shipped 62 days before DeepSeek V2.5, so benchmark comparisons should account for the intervening progress.

DeepSeek V2.5 has 236B parameters, while CodeGeeX4-ALL-9B has 9B. Context windows are 128k (DeepSeek V2.5) vs 128k (CodeGeeX4-ALL-9B).

On LiveCodeBench, DeepSeek V2.5 leads at 41.8% vs CodeGeeX4-ALL-9B at 28.5%. On HumanEval, DeepSeek V2.5 leads at 89% vs CodeGeeX4-ALL-9B at 82.3%.