DeepSeek Coder 4.0 30B

NEW HOT

DeepSeek AI · DeepSeek

Open SourceAPI

A highly capable code generation and understanding model, excelling in programming tasks across multiple languages.

TextCodeReasoning

Scores 85% on HumanEval, competitive across 4 standard benchmarks.

Data aggregated and editorially reviewed by TrendMing.

Benchmark Performance

MMLU79%
HumanEval85%
MATH58%
GPQA38%

Scores sourced from published evaluations. Higher = better.

Quick Specs

Context Window64K tokens
Parameters30B
Input pricingN/A (self-hosted)
Output pricingN/A (self-hosted)
Released2026-05-10