MatGPTQ: Accurate and Efficient Post-Training Matryoshka Quantization
Paper
• 2602.03537 • Published
• 3 •
1
Max Kleinegger
mkleinegger
·
AI & ML interests
None yet
Recent Activity
authored
a paper
8 days ago
MatGPTQ: Accurate and Efficient Post-Training Matryoshka Quantization upvoted a paper 8 days ago
QuEST: Stable Training of LLMs with 1-Bit Weights and Activations upvoted a paper 8 days ago
Bridging the Gap Between Promise and Performance for Microscaling FP4
Quantization Organizations
commented a paper 8 days ago