Michael Matheka

gamamima

·

AI & ML interests

None yet

Recent Activity

reacted to eaddario's post with 🔥 1 day ago

Experimental global target bits‑per‑weight quantization of IFM/K2-Horizon-7B and IFM/K2-Horizon-MoVA-36B-A4B. 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. Full benchmarks (PPL, KLD, ARC, GPQA, MMLU, etc.) and methodology in the model's card. https://huggingface.co/eaddario/K2-Horizon-7B-GGUF https://huggingface.co/eaddario/K2-Horizon-MoVA-36B-A4B-GGUF

reacted to eaddario's post with 🔥 1 day ago

Experimental global target bits‑per‑weight quantization of IFM/K2-Horizon-7B and IFM/K2-Horizon-MoVA-36B-A4B. 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. Full benchmarks (PPL, KLD, ARC, GPQA, MMLU, etc.) and methodology in the model's card. https://huggingface.co/eaddario/K2-Horizon-7B-GGUF https://huggingface.co/eaddario/K2-Horizon-MoVA-36B-A4B-GGUF

updated a model 4 days ago

gamamima/whisper-small-kikamba_lora_adapter_v1

View all activity

Organizations

spaces 2

Template Final Assignment

Run agent to answer questions and get scored

First Agent Template

models 1

gamamima/whisper-small-kikamba_lora_adapter_v1

Updated 4 days ago

datasets 1

gamamima/Kamba-ASR-Data-Subset-484H

Viewer • Updated 18 days ago • 161k • 354 • 1