Victor Nogueira

Felladrin

·

https://felladrin.com

felladrin

AI & ML interests

Models to run in the web browser

Recent Activity

updated a Space about 9 hours ago

Felladrin/MiniSearch-AI-Server

reacted to eaddario's post with 🔥 about 11 hours ago

Experimental global target bits‑per‑weight quantization of Qwen/Qwen3.8-27B, including MTP tensors in imatrix file. Unlike standard llama.cpp quantization that rely on fixed type heuristics (e.g., Q4_K_M), the Target BPW approach automatically optimizes per-tensor precision where it matters the most, and produces high quality models that meet a precise global size target. Key Advantages: - VRAM Maximization: Can generate high quality models sized exactly to fit hardware constraints (e.g., fitting the model into exactly 24GB VRAM). - Data-Driven Precision: Quantization mix is determined by actual weight error sensitivity rather than hardcoded rules, often yielding better PPL/KLD size trade-offs. - imatrix includes data for MTP/NextN layers. Full benchmarks (PPL, KLD, ARC, GPQA, MMLU, etc.) and methodology in the model's card. https://huggingface.co/eaddario/Qwen3.8-27B-GGUF

new activity about 20 hours ago

onnx-community/single-floppy-346k-merged-ONNX:Upload chat_template.jinja

View all activity

Organizations

Felladrin 's collections 5