# MoonshotAI/checkpoint-engine

Checkpoint-engine is a simple middleware to update model weights in LLM inference engines

Repository: https://github.com/MoonshotAI/checkpoint-engine
Canonical: https://ross.abutalabs.com/products/checkpoint-engine
Language: Python
License: MIT
License Family: permissive
Last push: 2026-08-12T05:59:00+00:00

## Health v2 (maintenance only)
Score: 80/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 97, release rhythm 91, longevity 25
- inputs: {"age_days": 359, "days_push": 21, "days_rel": 60, "gap_med": 12, "n_releases_24m": 14}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1005, forks 106 (observed 2026-09-03T02:15:06.726277+00:00)

## What it is
Checkpoint-engine is a lightweight Python middleware for updating model weights in-place across LLM inference engines, a critical step in reinforcement learning workflows. It provides efficient Broadcast and P2P weight transfer implementations, updating a 1-trillion-parameter model across thousands of GPUs in about 20 seconds.

## Use cases
- update LLM weights in inference engines during RL training
- synchronize model weights across thousands of GPUs
- push new weights to dynamically added inference instances
- broadcast sharded model weights from CPU memory to inference clusters
- integrate weight syncing between training and serving for RLHF

## When to choose
- you run reinforcement learning loops that need to refresh inference engine weights in-place
- you serve very large models (e.g., trillion-parameter MoE) across many GPUs and need fast weight sync
- you need to update weights on inference instances added mid-serving without disrupting existing workloads

## When to avoid
- you only need to save or load checkpoints to disk rather than live weight updates
- your inference engine is not supported/integrated with checkpoint-engine
- you need a full RL training framework rather than a weight-update middleware

## Facets
- artifact type: library
- maturity: active
- function: middleware, llm-inference, llm-training, reinforcement-learning, gpu-computing
- domain: large-language-models, machine-learning, reinforcement-learning, microservices, developer-tools
- platform: python
- tags: weight-update, parameter-server, rlhf, model-serving, broadcast, p2p-transfer, cuda-ipc, zeromq, gpu, linux, docker

## Member repositories
- MoonshotAI/checkpoint-engine (main) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:06.726277+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 2026-08-30T07:16:03.988799+00:00, confidence not recorded.
  - readme: https://github.com/MoonshotAI/checkpoint-engine (fetched 2026-09-03T02:15:06.726277+00:00, sha 7b552b335c97)
  - registry_pypi: https://pypi.org/pypi/checkpoint-engine/json (fetched 2026-08-29T13:14:04.754667+00:00, sha fcbea64f0f0a)
- Data as of 2026-08-30T08:39:29.467469+00:00.