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.
giannisanni/pulsar
View on GitHub ↗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.
pulsar grew +2.7% in the last 30 days, ranks 318th of 494 in Local Runtime / Inference Engine, last commit 53 days ago.
Star trajectory
83 days tracked+2.7% in 30d, above the Local Runtime / Inference Engine median of +1.7%.
Last commit 53 days ago, staler than the median of 5d.
How it compares in Local Runtime / Inference Engine
Ranked 318th of 494 by stars.
- Language
- Rust
- License
- Other
- Contributors
- 7
- Forks
- 28
- Watchers
- 2
- Open issues
- 6
- Releases
- 0
- Created
- 2026-07-13
- Stars this week
- +1
You've seen where pulsar stands today.
Get the moves before everyone else: the 5 Local Runtime / Inference Engine repos climbing fastest, every Tuesday.
Free weekly digest. Double opt-in, unsubscribe anytime.