marlin

IST-DASLab

FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.

kernel

View on GitHub

1.1k Stars

86 Forks

1.1k Watchers

Python Language

apache-2.0 License

100 SrcLog Score

Cost to Build

$26.9K

Market Value

$83.7K

How is this calculated?

Growth over time

3 data points · 2026-04-08 → 2026-04-22

Stars Forks Watchers

💬

How do you feel about this project?

Ask AI about marlin

Question copied to clipboard

What is the IST-DASLab/marlin GitHub project? Description: "FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.". Written in Python. Explain what it does, its main use cases, key features, and who would benefit from using it.

Question is copied to clipboard — paste it after the AI opens.

How to clone marlin

Clone via HTTPS

git clone https://github.com/IST-DASLab/marlin.git

Clone via SSH

[email protected]:IST-DASLab/marlin.git

Download ZIP

Download master.zip

Found an issue?

Report bugs or request features on the marlin issue tracker:

Open GitHub Issues

Similar to marlin

gvisor raspberry-pi-os reactos build-linux cilium spectre-meltdown-checker toaruos linux-insides-zh windows-kernel-exploits syzkaller linux-kernel-exploits rt-thread blog_os gophernotes pmbootstrap tock build simplefs u-root kernelpop SpecuCheck xos ksm sparkmagic PowerNex frakti Diamorphine The_Holy_Book_of_X86 jupyter-c-kernel kAFL