microsoft/MInference
[NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention, which reduces inference latency by up to 10x for pre-filling
- Stars
- 1.2k
- Forks
- 84
- Open issues
- 87
- Last push
- 12d ago
About MInference
MInference is a free, open-source AI inference tool by microsoft, written mainly in Python. [NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention, which reduces inference latency by up to 10x for pre-filling. It has 1.2k GitHub stars and 84 forks, and was last updated 12d ago. It's released under the MIT license, a standard open-source license.
- License
- MIT
- Language
- Python
- Stars
- 1,229
- Last update
- 12d ago
- Runs on Mac
- Not stated
- GPU
- Not stated
- Paid API key
- Not required
- Self-hostable
- Yes
Similar free tools
TensorRT-LLM15k stars · TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT
BentoML8.9k stars · The easiest way to serve AI apps and models - Build Model Inference APIs, Job queues, LLM apps, Multi-model pipelines, and more!
vllm95k stars · A high-throughput and memory-efficient inference and serving engine for LLMs
distributed-llama3.1k stars · Distributed LLM inference. Connect home devices into a powerful cluster to accelerate LLM inference. More devices means faster inference.
kimi-k3-in-c9.2k stars · A 2.78-trillion-parameter Kimi K3 running inference on a single CPU in 8.24 GB of RAM. Portable C99: no BLAS, no framework, no GPU.
MInference FAQ
Is MInference free?
Yes. MInference is open source and free to use. It's released under the MIT license, a standard open-source license.
Does MInference run on a Mac?
The project doesn't say. Check its installation guide before you start.
Does MInference need a GPU?
The project doesn't state hardware needs. AI models generally run much faster with a GPU.
Is MInference still maintained?
Yes. It was last updated 12d ago and has 87 open issues on GitHub.
What are alternatives to MInference?
Similar free, open-source tools include TensorRT-LLM, BentoML, OpenLLM, vllm.