ddh0

ddh0

·

ddh0

AI & ML interests

None yet

Recent Activity

liked a model 29 minutes ago

sophosympatheia/Glistening-Gem-31B-v2.2

liked a model about 20 hours ago

bartowski/Ling-3.0-flash-GGUF

reacted to eaddario's post with ❤️ about 21 hours ago

Experimental JQK and global target bits‑per‑weight quantization of Nanbeige/Nanbeige4.3-33B. Unlike standard llama.cpp quantization that relies 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: - Better model quality: using JQK types can improve KLD between 10% to 50% compared to standard QK types, at the same bpw. - 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. Information about JQK types: https://huggingface.co/eaddario/JQ2_K-JQ6_K-Test-Models Full benchmarks (PPL, KLD, ARC, GPQA, MMLU, etc.) and methodology in the model's cards. JQK Quantization: https://huggingface.co/eaddario/Nanbeige4.3-33B-JQK Global target BPW: https://huggingface.co/eaddario/Nanbeige4.3-33B-GGUF

View all activity

Organizations

ddh0 's datasets

None public yet