# OpenBMB/CPM-Bee

百亿参数的中英文双语基座大模型

Repository: https://github.com/OpenBMB/CPM-Bee
Canonical: https://ross.abutalabs.com/products/cpm-bee
Language: Python
License Family: other
Last push: 2026-07-07T14:05:53+00:00

## Health v2 (maintenance only)
Score: 71/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 91, release rhythm 35, longevity 87
- inputs: {"age_days": 1226, "days_push": 57, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2406, forks 176 (observed 2026-08-28T04:06:44.157748+00:00)

## What it is
CPM-Bee is a fully open-source, commercially usable 10-billion-parameter bilingual (Chinese-English) foundation large language model trained on over a trillion tokens of curated corpus. The repository provides model weights, inference code, and fine-tuning tutorials built on the OpenBMB ecosystem of training and compression tools.

## Use cases
- run a 10B bilingual Chinese-English LLM locally
- fine-tune a foundation model on domain-specific data
- build chat or tool-using agents on an open base model
- perform text generation, translation, QA, and fill-in-the-blank tasks
- adapt a base model for retrieval-augmented applications like WebCPM
- use a commercially licensed open-source LLM for research or products

## When to choose
- you need a strong open bilingual Chinese-English base model with commercial rights
- you want to fine-tune or align a foundation model on your own data
- you are building on the OpenBMB ecosystem (BMTrain, model compression, inference tools)

## When to avoid
- you need a production-ready instruction-tuned chat model out of the box
- you lack GPU hardware capable of serving a 10B parameter model
- you require a permissive license without restrictions - the repo lists no explicit license
- you need active development - updates have been infrequent since 2023

## Facets
- artifact type: library
- maturity: maintenance
- function: llm-inference, llm-training, machine-learning, nlp
- domain: large-language-models, artificial-intelligence, deep-learning
- platform: python
- tags: bilingual, chinese-english, foundation-model, transformer, openbmb, pretrained-model, commercial-use, natural-language-processing, gpu, linux

## Member repositories
- OpenBMB/CPM-Bee (main) score 71

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:44.157748+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 2026-08-30T02:34:05.762876+00:00, confidence not recorded.
  - readme: https://github.com/OpenBMB/CPM-Bee (fetched 2026-08-28T04:06:44.157748+00:00, sha 0967e2ccbb70)
- Data as of 2026-08-30T08:39:29.467469+00:00.