Malandrino PRO

Pier-Jean

·

AI & ML interests

Document parsing Quantization

Recent Activity

authored a paper 5 days ago

Unfolding the Leech Lattice: Fused Multi-Shell Decoding and VRAM Layouts for 2-Bit LLM Weights

published an article 9 days ago

I put a 24 dimensional lattice in a CUDA kernel to run Qwen3-4B in 2.6 GB

updated a model 20 days ago

Pier-Jean/Qwen3-4B-LLVQ-2bit

View all activity

Organizations

Articles 3

Article

I put a 24 dimensional lattice in a CUDA kernel to run Qwen3-4B in 2.6 GB

Article

Docling Studio — Open-Source Visual Inspection for Docling Pipelines

View all Articles

Papers 1

arxiv:2609.02652

spaces 1

Docling Studio

Generate and edit custom documents online

models 5

Pier-Jean/Qwen3-4B-LLVQ-2bit

Updated 20 days ago

Pier-Jean/qwen3-8b-awq-deq

8B • Updated Aug 7 • 7

Pier-Jean/qwen3-4b-awq-deq

4B • Updated Aug 6 • 11

Pier-Jean/scubsql-coder

Updated Aug 4, 2024

Pier-Jean/autotrain-psamz-ydtct

60.5M • Updated Aug 4, 2024 • 7

datasets 0

None public yet