# InternLM/HuixiangDou

HuixiangDou: Overcoming Group Chat Scenarios with LLM-based Technical Assistance

Repository: https://github.com/InternLM/HuixiangDou
Canonical: https://ross.abutalabs.com/products/huixiangdou
Homepage: https://openxlab.org.cn/apps/detail/tpoisonooo/huixiangdou-web
Language: Python
License: BSD-3-Clause
License Family: permissive
Topics: chatbot, llm, rag, dsl, lark, pipeline, robot, wechat, application, multimodal, image-retrieval, assistant, assistant-chat-bots, group-chat
Last push: 2025-11-24T11:17:29+00:00

## Health v2 (maintenance only)
Score: 46/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 53, release rhythm 25, longevity 69
- inputs: {"age_days": 979, "days_push": 282, "days_rel": 289, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2499, forks 179 (observed 2026-08-28T04:06:56.873898+00:00)

## What it is
HuixiangDou is an LLM-based professional knowledge assistant designed for group chat scenarios, using a three-stage pipeline of preprocess, rejection, and response to answer technical questions without flooding conversations. It requires no model training, runs on CPU-only configurations, and ships Web, Android, and pipeline source code for WeChat, Lark, and other integrations.

## Use cases
- deploy an AI assistant in a WeChat group chat
- answer technical questions from a knowledge base without spamming the group
- build a RAG chatbot over project documentation
- stream chat responses against a code repository
- add a Q&A bot to Lark or Discord groups
- run a knowledge assistant on CPU-only hardware
- retrieve answers from documents including images

## When to choose
- you need a group-chat-aware assistant that filters noise and only answers relevant questions
- you want a no-training RAG pipeline deployable on modest CPU hardware
- you need WeChat, Lark, or web integrations with commercial-friendly licensing (BSD-3-Clause)

## When to avoid
- you need a general-purpose conversational chatbot rather than a domain knowledge assistant
- you work outside computer science and want GraphRAG-style retrieval, where HuixiangDou2 may fit better
- you require fully managed cloud hosting with no self-hosting

## Facets
- artifact type: application
- maturity: active
- function: rag, chatbot, llm-inference, nlp, image-processing, search-engine, webhook
- domain: large-language-models, chatbots, artificial-intelligence, developer-tools
- platform: python, self-hosted
- tags: group-chat, wechat-bot, lark, knowledge-assistant, rejection-pipeline, multimodal-retrieval, no-training, retrieval-augmented-generation, natural-language-processing, docker, web-server, android

## Member repositories
- InternLM/HuixiangDou (main) score 46

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:56.873898+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 2026-08-30T02:26:57.841990+00:00, confidence not recorded.
  - readme: https://github.com/InternLM/HuixiangDou (fetched 2026-08-28T04:06:56.873898+00:00, sha e105d531c525)
  - homepage: https://openxlab.org.cn/apps/detail/tpoisonooo/huixiangdou-web (fetched 2026-08-29T10:09:33.067440+00:00, sha 39ac88edc400)
  - registry_pypi: https://pypi.org/pypi/huixiangdou/json (fetched 2026-08-29T10:09:33.076477+00:00, sha 86d4b463a72a)
- Data as of 2026-08-30T08:39:29.467469+00:00.