turboquant-pytorch
From-scratch PyTorch implementation of Google's TurboQuant (ICLR 2026) for LLM KV cache compression. 5x compression at 3-bit with 99.5% attention fidelity.
tonbistudio/turboquant-pytorch
View on GitHub ↗From-scratch PyTorch implementation of Google's TurboQuant (ICLR 2026) for LLM KV cache compression. 5x compression at 3-bit with 99.5% attention fidelity.
turboquant-pytorch grew +0.3% in the last 30 days, ranks 12th of 101 in Model Optimization, last commit 182 days ago.
Star trajectory
113 days tracked2026-07-021k stars today2026-10-23
Momentum
+0.3% in 30d, below the Model Optimization median of +0.6%.
Maintenance
Last commit 182 days ago, staler than the median of 51d.
How it compares in Model Optimization
turboquant-pytorch
1k
stars
+0.3% 30d
commit 182 days ago
Model Optimization median
266
stars
+0.6% 30d
commit ~51d ago
Leader · kimi-k3-in-c
9.2k
stars
+9.6% 30d
commit 21 days ago
Ranked 12th of 101 by stars.
intel/auto-round1.7k-637RWKV/rwkv.cpp1.6k-539
tonbistudio/turboquant-pytorch1k
sophgo/tpu-mlir1k+40OpenGVLab/OmniQuant964+85- Language
- Python
- License
- MIT License
- Contributors
- 1
- Forks
- 140
- Watchers
- 4
- Open issues
- 13
- Releases
- 0
- Created
- 2026-03-25
- Stars this week
- +0
You've seen where turboquant-pytorch stands today.
Get the moves before everyone else: the 5 Model Optimization repos climbing fastest, every Tuesday.
Free weekly digest. Double opt-in, unsubscribe anytime.