Home Projects KVSplit
KVSplit

KVSplit

by dipampaul17 · GitHub

Run larger LLMs with longer contexts on Apple Silicon by using differentiated precision for KV cache quantization. KVSplit enables 8-bit keys & 4-bit values, reducing memory by 59% with <1% quality loss. Includes benchmarking, visualization, and one-command se

View on GitHub
⭐ Stars
360
🍴 Forks
13
📜 License
License not specified
📅 Created
2025
🔄 Last commit
1 yr ago
🏷️ Category
apple-silicon
💻 Language
🖥️ Self-hostable
Likely
You maintain this project?

Claim its page: indexed whatever its rank, translated into six languages, and enriched with what you write yourself.

Claim this page →
KVSplit — GitHub preview card
📈 Star history
361360
2026-07-202026-07-21
📄 About

Run larger LLMs with longer contexts on Apple Silicon by using differentiated precision for KV cache quantization. KVSplit enables 8-bit keys & 4-bit values, reducing memory by 59% with <1% quality loss. Includes benchmarking, visualization, and one-command se

Frequently asked questions

What is KVSplit?

Run larger LLMs with longer contexts on Apple Silicon by using differentiated precision for KV cache quantization. KVSplit enables 8-bit keys & 4-bit values, reducing memory by 59% with

Is KVSplit open source?

KVSplit is an open-source project.

Is KVSplit free?

Yes. KVSplit is free and open source — you can use, modify and self-host it.

🏅 Maintainer of this project?
OpenSourceAI badge — KVSplit

Add this live badge to your README — your GitHub stars and directory rank, refreshed daily.

[![OpenSourceAI](https://opensourceai.tech/badge.php?tool=dipampaul17-kvsplit)](https://opensourceai.tech/project/dipampaul17-kvsplit.html)
More badge options →
🧬 Related projects🧬 View the DNA map →