Checkpoint Engine
checkpoint-engine is a lightweight, decoupling and efficient weight update middleware
Model DeploymentMLOps & Experiment TrackingOpen Source
About
Checkpoint-engine is a simple middleware to update model weights in LLM inference engines -- a critical step in reinforcement learning. We provide an efficient and lightweight implementation for inplace weight update: updating our Kimi-K2 model (1 Trillion parameters) across thousands of GPUs takes about 20s.
Open Source Health
- Stars
- 1,005
- Forks
- 107
- License
- MIT
- Last commit
- 1 months ago
Python
Alternatives to Checkpoint Engine
CHaiDNNby XilinxHLS based Deep Neural Network Accelerator Library for Xilinx Ultrascale+ MPSoCs- SoulX-FlashTalkby Soul AI LabSoulX-FlashTalk is the first 14B model to achieve sub-second start-up latency (0.87s) while maintaining a real-time throughput of 32 FPS on an 8xH800 node.
Ts Patternby Gabriel VergnaudThe exhaustive Pattern Matching library for TypeScript, with smart type inference.
Nobodywhoby NobodyWhoNobodyWho is an inference engine that lets you run LLMs locally and efficiently on any device.
ds4by Salvatore SanfilippoDeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm
Resources & Links
Related Categories
Using Checkpoint Engine?
Track its cost next to the rest of your stack and get a reminder before it renews.
Add to my stackVendor
Moonshot AI
Open-source projects on GitHub: Kimi-Dev, MoonEP and Checkpoint Engine
