FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.
-
Updated
Sep 4, 2024 - Python
#
4bit
Here are 9 public repositories matching this topic...
QuantLLM is a Python library designed for developers, researchers, and teams who want to fine-tune and deploy large language models (LLMs) efficiently using 4-bit and 8-bit quantization techniques.
python quantum pypi transformers torch python3 pytorch quantization 8bit 4bit pypi-packages python-lambda pypi-package huggingface llm
-
Updated
Apr 29, 2026 - Python
Supporting code for "LLMs for your iPhone: Whole-Tensor 4 Bit Quantization"
-
Updated
Mar 31, 2024 - Python
Mixed-precision quantization scheme (16/8/4bit mixed quantization) for the Wan2.2-Animate-14B model. Compresses the original 35GB base model to 17GB, balancing inference performance and model size.
quantization 8bit 4bit 16bit model-compression image2video wan2 wan2-2-animate wan2-animate mixed-precision-quantization 14b-model
-
Updated
Dec 22, 2025 - Python
a 4 bit TTL computer
-
Updated
Oct 1, 2024 - Python
ACE-Step 1.5 XL Optimized Fork: 4-Bit (INT4) Windows support + RTX 2080 Ti (Turing) stability fixes. Turing architecture compatible. Runs XL SFT on 11GB VRAM without OOM.
-
Updated
Apr 23, 2026 - Python
Improve this page
Add a description, image, and links to the 4bit topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the 4bit topic, visit your repo's landing page and select "manage topics."