# MoonshotAI/Kimi-Linear

Repository: https://github.com/MoonshotAI/Kimi-Linear
Canonical: https://ross.abutalabs.com/products/kimi-linear
License: MIT
License Family: permissive
Last push: 2025-11-17T09:36:48+00:00

## Health v2 (maintenance only)
Score: 40/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 52, release rhythm 35, longevity 22
- inputs: {"age_days": 308, "days_push": 289, "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 1588, forks 87 (observed 2026-08-28T04:05:08.003698+00:00)

## What it is
Kimi Linear is a hybrid linear attention architecture (Kimi Delta Attention, based on Gated DeltaNet) released by Moonshot AI with 48B-parameter MoE model checkpoints (Base and Instruct) supporting 1M-token contexts. The repo provides the model implementation, KDA kernel integration with flash-linear-attention, and usage instructions for inference and fine-tuning.

## Use cases
- run a long-context LLM with 1M token window
- use linear attention models instead of full attention transformers
- fine-tune a hybrid linear attention MoE model
- reduce KV cache memory for long-context inference
- benchmark linear attention vs full attention decoding throughput
- experiment with Kimi Delta Attention kernels

## When to choose
- you need very long context (up to 1M tokens) with high decoding throughput
- you want to research or build on linear attention / finite-state RNN memory architectures
- you want lower KV cache memory usage in LLM serving

## When to avoid
- you need a small model that fits on consumer hardware (48B total params)
- you need broad ecosystem/tooling support of mainstream transformer models
- you require short-context tasks where full attention models are equally good and better supported

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, llm-inference, llm-training
- domain: large-language-models, deep-learning, artificial-intelligence
- platform: python
- tags: linear-attention, kimi-delta-attention, kda, long-context, moe, flash-linear-attention, model-checkpoints, transformers, gpu, linux

## Member repositories
- MoonshotAI/Kimi-Linear (main) score 40

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:08.003698+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 2026-08-30T03:54:59.196020+00:00, confidence not recorded.
  - readme: https://github.com/MoonshotAI/Kimi-Linear (fetched 2026-08-28T04:05:08.003698+00:00, sha 5dcdf7a89b28)
- Data as of 2026-08-30T08:39:29.467469+00:00.