# meituan-longcat/LongCat-Flash-Chat

Repository: https://github.com/meituan-longcat/LongCat-Flash-Chat
Canonical: https://ross.abutalabs.com/products/longcat-flash-chat
License: MIT
License Family: permissive
Last push: 2026-06-23T12:07:22+00:00

## Health v2 (maintenance only)
Score: 58/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 89, release rhythm 35, longevity 26
- inputs: {"age_days": 368, "days_push": 71, "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 1365, forks 74 (observed 2026-08-28T04:04:31.177389+00:00)

## What it is
LongCat-Flash-Chat is the official repository for Meituan's LongCat-Flash, a 560-billion-parameter Mixture-of-Experts (MoE) foundation language model with dynamic parameter activation (~27B average per token). The repo provides model weights, technical documentation, and inference guidance for this non-thinking chat model optimized for agentic tasks.

## Use cases
- download and run a large open-weight MoE language model
- serve a chat model for agentic workflows
- evaluate LongCat-Flash against other leading LLMs
- deploy efficient LLM inference with high tokens-per-second throughput
- fine-tune or build on top of an open foundation model
- study MoE architecture with shortcut-connected design

## When to choose
- you need a powerful open-weight MoE chat model with strong agentic capabilities
- you want cost-efficient inference with dynamic computation activation
- you need a non-thinking foundation model for building agents

## When to avoid
- you lack GPU infrastructure for a 560B-parameter model
- you need a small model for edge or local laptop use
- you need a reasoning/thinking model rather than a foundation chat model

## Facets
- artifact type: learning-resource
- maturity: active
- function: llm-inference, machine-learning, deep-learning
- domain: large-language-models, artificial-intelligence
- platform: python
- tags: mixture-of-experts, foundation-model, open-weights, huggingface, meituan, longcat-flash, agentic-tasks, model-weights, ai-agents, gpu, linux, docker

## Member repositories
- meituan-longcat/LongCat-Flash-Chat (main) score 58

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:31.177389+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 2026-08-30T04:41:16.650327+00:00, confidence not recorded.
  - readme: https://github.com/meituan-longcat/LongCat-Flash-Chat (fetched 2026-08-28T04:04:31.177389+00:00, sha 60f1192e37d5)
- Data as of 2026-08-30T08:39:29.467469+00:00.