Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
79.1
TFLOPS
Patrick Devaney
patrickbdevaney
7
15
336
Follow
laooopooo's profile picture
Rebis's profile picture
KatyaKnw's profile picture
13 followers
·
38 following
patrickbdevaney
patrickbdevaney
AI & ML interests
mamba s4, mixtral 8x-7b, candle, pytorch
Recent Activity
liked
a model
1 minute ago
Altworld/Hemmingway-1
reacted
to
onekq
's
post
with 👀
5 minutes ago
My takes on Jev 1. Very likely a small model. You can certainly pretrain, but I would grab an existing base model, say Qwen 3 class 2. The new RL method is a breakthrough, classification doesn't need to align with human preferences 3. The new output is an overstatement. It's just a new LM head. Of course autoregressive decoding can be used for classification: it takes just a few tokens to express the output. Think twice: are you sure classification doesn't need few-shot, CoT, or reasoning? All of these depend on auto-regressiveness 4. It carves out a market already existing, which is now served by oversized LLMs (hence overpaid), e.g. LLM as judge, labeling 5. Jevons effect will kick in, promoting more modeling efforts for small budget teams. It might even accelerate RSI
reacted
to
eaddario
's
post
with 🔥
5 days ago
Experimental global target bits‑per‑weight quantization of openbmb/MiniCPM5-1B and openbmb/MiniCPM5-2B. 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/MiniCPM5-1B-GGUF https://huggingface.co/eaddario/MiniCPM5-2B-GGUF
View all activity
Organizations
patrickbdevaney
's activity
All
Models
Datasets
Spaces
Buckets
Papers
Collections
Community
Posts
Upvotes
Likes
Articles
New activity in
patrickbdevaney/GLM-5.3-Flash-REAP50-GGUF
19 days ago
Request: NVFP4 Version
3
#2 opened 19 days ago by
wa999
New activity in
patrickbdevaney/GLM-5.3-Flash-REAP50-GGUF
23 days ago
Excited to test :)
❤️
2
5
#1 opened 23 days ago by
Darkknight535
New activity in
patrickbdevaney/GLM-5.3-Flash-REAP50-FP8
25 days ago
GGUF Quants
2
#1 opened 25 days ago by
Darkknight535
New activity in
patrickbdevaney/Qwen3.5-122B-A10B-NVFP4-vllm-docker-jetson-agx-thor
4 months ago
Finally Qwen3.5-122B-A10B-NVFP4 working on Thor!
❤️
1
1
#1 opened 7 months ago by
pastoriomarco
New activity in
LeRobot-worldwide-hackathon/submissions
over 1 year ago
Uploading demo: TEAM-183-SimuRL.mp4
#195 opened over 1 year ago by
patrickbdevaney
New activity in
patrickbdevaney/codellama7b-instruct-solidity-gguf
over 2 years ago
Create AddressAssociationAttesterRegistry.sol
#2 opened over 2 years ago by
laksc
Create README.md
#1 opened over 2 years ago by
laksc