# deepseek-ai/DeepSeek-Coder-V2

DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence

Repository: https://github.com/deepseek-ai/DeepSeek-Coder-V2
Canonical: https://ross.abutalabs.com/products/deepseek-coder-v2
License: MIT
License Family: permissive
Last push: 2025-11-11T06:44:45+00:00

## Health v2 (maintenance only)
Score: 47/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 51, release rhythm 35, longevity 57
- inputs: {"age_days": 810, "days_push": 295, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7007, forks 1111 (observed 2026-08-28T04:09:53.228048+00:00)

## What it is
DeepSeek-Coder-V2 is an open-weight Mixture-of-Experts code language model family released by DeepSeek AI, with weights available on Hugging Face under a model license (code under MIT). The repository provides model downloads, evaluation results, API access, and instructions for running the models locally.

## Use cases
- generate code from natural language descriptions
- run a local code completion model
- fill in the middle for code infilling
- evaluate code intelligence benchmarks
- serve a code LLM via an API
- fine-tune a code model for my repo

## When to choose
- you need an open-weight model specialized in code intelligence
- you want to run a strong code LLM locally or on your own GPU infrastructure
- you need code generation, completion, or infilling across many programming languages

## When to avoid
- you need a small model for low-resource devices
- you only want a hosted API without managing model weights
- you need general chat rather than code-focused capabilities

## Facets
- artifact type: library
- maturity: active
- function: llm-inference, machine-learning, deep-learning
- domain: large-language-models, artificial-intelligence, developer-tools
- platform: python, cross-platform
- tags: code-generation, open-weights, mixture-of-experts, huggingface, model-weights, gpu

## Member repositories
- deepseek-ai/DeepSeek-Coder-V2 (main) score 47

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:53.228048+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:40:44.442221+00:00, confidence not recorded.
  - readme: https://github.com/deepseek-ai/DeepSeek-Coder-V2 (fetched 2026-08-28T04:09:53.228048+00:00, sha df1465bbb8da)
- Data as of 2026-08-30T08:39:29.467469+00:00.