nvidia/NVIDIA-Nemotron-3.5-ASR-Streaming-Multilingual-0.6b
Updated • 34 • 17
Zdenek Slavik
slavikz
·
AI & ML interests
None yet
Recent Activity
new activity about 18 hours ago
nvidia/nemotron-speech-streaming-en-0.6b:Multilingual version planned? liked a model 1 day ago
nvidia/NVIDIA-Nemotron-3.5-ASR-Streaming-Multilingual-0.6b updated a model 6 days ago
inferRouter/Qwen3.6-35B-A3B-NVFP4Organizations
Multilingual version planned?
12
#2 opened 4 months ago
by 
fosple
liked a
model 1 day ago
updated a
model 6 days ago
published a
model 6 days ago
updated a
model 7 days ago
reacted to eaddario's post with 🚀 15 days ago
Post
3163
Experimental global target bits‑per‑weight quantization of Qwen/Qwen3.6-27B and Qwen/Qwen3.6-35B-A3B.
Unlike standard llama.cpp quantizations that rely on fixed type heuristics (e.g., Q4_K_M), the Target BPW approach optimizes per-tensor precision where it matters the most, and produces high quality models that meet a precise global file 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 models' cards.
eaddario/Qwen3.6-27B-GGUF
eaddario/Qwen3.6-35B-A3B-GGUF
Unlike standard llama.cpp quantizations that rely on fixed type heuristics (e.g., Q4_K_M), the Target BPW approach optimizes per-tensor precision where it matters the most, and produces high quality models that meet a precise global file 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 models' cards.
eaddario/Qwen3.6-27B-GGUF
eaddario/Qwen3.6-35B-A3B-GGUF
updated a
model 19 days ago
published a
model 19 days ago
updated a
model 19 days ago
published a
model 21 days ago
published a
model 22 days ago
Why is this 4bit version has a 32.7 GB size?
➕ 3
20
#3 opened about 2 months ago
by 
alexcardo
Create preprocessor_config.json
#1 opened 4 months ago
by slavikz
