Skip to content
P

Pulsar

SSD-streaming inference engine for giant MoE models (Rust + CUDA)

by giannisanniFree

Launch
#4734in AI
0In stores

Description

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…

Ratings & Reviews

No ratings yet

Be the first to rate Pulsar.

People also visit

F

Fox

A local LLM server built for concurrent work

AI
Launch
P

Project Maya

Run GLM-5.3-Flash (321B MoE) privately on your own GPU(s)…

AI
Launch
G

Ggrun

llama.cpp/ik_llama.cpp launcher: loads big MoE models across mismatched

AI
Launch
V

VoiceStudio

VoiceStudio is the open-source, fully-local ElevenLabs alternative

AI
Launch
L

Llamafile

Distribute and run LLMs with a single file

AI
Launch
V

Viseron

Self-hosted, local only NVR and AI Computer Vision software

AI
Launch

Pulsar FAQ

What is 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 attention and hot experts where they fit.

Is Pulsar free?

Yes. Pulsar is free to use.

Do I need to download Pulsar?

No. Pulsar runs in your browser at github.com, so there's nothing to install.

Who makes Pulsar?

Pulsar is made by giannisanni.

What are the best Pulsar alternatives?

Popular alternatives to Pulsar include Fox, Project Maya and Ggrun. See all Pulsar alternatives

Developer Stuff

Badges