2 2 27

Yakov Saparov

Outrun32

AI & ML interests

ML Engineering

Recent Activity

liked a model about 1 month ago

timm/tf_efficientnetv2_b0.in1k

liked a model 2 months ago

ali-vilab/In-Context-LoRA

View all activity

Organizations

Outrun32's activity

liked a model about 1 month ago

timm/tf_efficientnetv2_b0.in1k

Image Classification • Updated Apr 27, 2023 • 866k • 2

liked a model 2 months ago

ali-vilab/In-Context-LoRA

Text-to-Image • Updated 22 days ago • 96.9k • • 513

liked 2 models 3 months ago

stabilityai/stable-diffusion-3.5-large

Text-to-Image • Updated Oct 22, 2024 • 124k • • 1.82k

aleksa-codes/flux-ghibsky-illustration

Text-to-Image • Updated Nov 14, 2024 • 25.8k • • 201

New activity in mlabonne/BigLlama-3.1-1T-Instruct 5 months ago

Recommended hardware?

#1 opened 5 months ago by

sdalemorrey

liked a model 6 months ago

OzzyGT/RealVisXL_V4.0_inpainting

Image-to-Image • Updated Apr 20, 2024 • 3.21k • 12

liked a Space 6 months ago

Sleeping

🏞️

PixArt 900M 1024px Base Model

liked 2 models 7 months ago

OpenGVLab/InternVL-Chat-V1-5

Image-Text-to-Text • Updated 21 days ago • 2.02k • 405

xinsir/controlnet-scribble-sdxl-1.0

Text-to-Image • Updated Jul 9, 2024 • 12.7k • 226

upvoted a paper 8 months ago

Your Transformer is Secretly Linear

Paper • 2405.12250 • Published May 19, 2024 • 149

reacted to mlabonne's post with 🚀 9 months ago

Post

9264

⚡ AutoQuant

AutoQuant is the evolution of my previous AutoGGUF notebook (https://colab.research.google.com/drive/1P646NEg33BZy4BfLDNpTz0V0lwIU3CHu). It allows you to quantize your models in five different formats:

- GGUF: perfect for inference on CPUs (and LM Studio)
- GPTQ/EXL2: fast inference on GPUs
- AWQ: super fast inference on GPUs with vLLM (https://github.com/vllm-project/vllm)
- HQQ: extreme quantization with decent 2-bit and 3-bit models

Once the model is converted, it automatically uploads it on the Hugging Face Hub. To quantize a 7B model, GGUF only needs a T4 GPU, while the other methods require an A100 GPU.

Here's an example of a model I quantized using HQQ and AutoQuant: mlabonne/AlphaMonarch-7B-2bit-HQQ

I hope you'll enjoy it and quantize lots of models! :)

💻 AutoQuant: https://colab.research.google.com/drive/1b6nqC7UZVt8bx4MksX7s656GXPM-eWw4