# deepseek-ai/DeepSeek-LLM

DeepSeek LLM: Let there be answers

Repository: https://github.com/deepseek-ai/DeepSeek-LLM
Canonical: https://ross.abutalabs.com/products/deepseek-llm
Homepage: https://chat.deepseek.com/
Language: Makefile
License: MIT
License Family: permissive
Last push: 2024-02-04T12:22:16+00:00

## Health v2 (maintenance only)
Score: 27/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 72
- inputs: {"age_days": 1008, "days_push": 941, "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 7257, forks 1293 (observed 2026-08-28T04:09:57.861443+00:00)

## What it is
DeepSeek LLM is a family of open-source large language models (7B and 67B, Base and Chat variants) trained from scratch on 2 trillion tokens of English and Chinese text. The repository provides model weights, quick-start inference code, and evaluation details for research use.

## Use cases
- run a 67B open-source chat model locally
- fine-tune a bilingual English-Chinese LLM
- benchmark open LLMs on coding and math tasks
- research training recipes for large language models
- self-host an alternative to Llama2 70B
- evaluate LLM performance on the Hungarian national exam

## When to choose
- you need strong open-weight models with good Chinese language support
- you want to study or reproduce LLM training at scale
- you need a self-hosted chat model for coding and math

## When to avoid
- you need a maintained production framework rather than model weights
- you lack GPU resources for 7B-67B parameter models
- you need the newest DeepSeek models (this repo is superseded by later releases)

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-inference, llm-training, machine-learning
- domain: large-language-models, artificial-intelligence
- platform: python
- tags: open-weights, bilingual, chinese-english, huggingface, model-weights, natural-language-processing, gpu, linux

## Member repositories
- deepseek-ai/DeepSeek-LLM (main) score 27

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:57.861443+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 2026-08-29T17:39:43.034563+00:00, confidence not recorded.
  - readme: https://github.com/deepseek-ai/DeepSeek-LLM (fetched 2026-08-28T04:09:57.861443+00:00, sha a4c3880b1d56)
- Data as of 2026-08-30T08:39:29.467469+00:00.