giannisanni/pulsar
SSD-streaming inference engine for giant MoE models (Rust + CUDA). GLM 5.2 743B at 2 tok/s and Hy3 295B at 7 tok/s on two consumer 16GB GPUs. Zero-config multi-GPU: measures PCIe bandwidth, places att
- Stars
- 230
- Forks
- 28
- Open issues
- 6
- Last push
- 1mo ago
About pulsar
pulsar is a free, open-source AI inference tool by giannisanni, written mainly in Rust. SSD-streaming inference engine for giant MoE models (Rust + CUDA). GLM 5.2 743B at 2 tok/s and Hy3 295B at 7 tok/s on two consumer 16GB GPUs. Zero-config multi-GPU: measures PCIe bandwidth, places att. It has 230 GitHub stars and 28 forks, and was last updated 1mo ago. GitHub doesn't detect a standard license, so check the repository before using it commercially. According to the project, it works on macOS.
- License
- Not detected
- Language
- Rust
- Stars
- 230
- Last update
- 1mo ago
- Runs on Mac
- Yes (as stated by the project)
- GPU
- Not stated
- Paid API key
- Not required
- Self-hostable
- Not stated
Similar free tools
vllm95k stars · A high-throughput and memory-efficient inference and serving engine for LLMs
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
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.
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!
ai-infra-book6.9k stars · 《深入理解 AI Infra：量化分析与系统设计》（李博杰 著）开源书稿：从硬件约束和模型架构出发，量化推导 LLM 推理与训练系统设计。含全书正文、PDF、配套计算工具与实验
pulsar FAQ
Is pulsar free?
Yes. pulsar is open source and free to use. GitHub doesn't detect a standard license, so check the repository before using it commercially.
Does pulsar run on a Mac?
Yes. According to the project, pulsar works on macOS.
Does pulsar need a GPU?
The project doesn't state hardware needs. AI models generally run much faster with a GPU.
Is pulsar still maintained?
It was last updated 1mo ago. Check recent issues on GitHub to judge how active it is.
What are alternatives to pulsar?
Similar free, open-source tools include vllm, TensorRT-LLM, kimi-k3-in-c, BentoML.