alkinun's picture

alkinun

AtAndDev

·

AI & ML interests

LLMs, Alignment, Merging, Unsloth, DPO, SFT, ORPO, SPIN..

Recent Activity

reacted to MonsterMMORPG's post with 🤗 about 6 hours ago

It is now possible to generate 16 Megapixel (4096x4096) raw images with SANA 4K model using under 8GB VRAM, 4 Megapixel (2048x2048) images using under 6GB VRAM, and 1 Megapixel (1024x1024) images using under 4GB VRAM thanks to new optimizations 13 January 2024 Update Installers : https://www.patreon.com/posts/from-nvidia-labs-116474081 New 4K Tutorial Video : https://youtu.be/GjENQfHF4W8 Now the APP will use Diffusers Pipeline and it has huge VRAM optimizations You need to reinstall The models will be downloaded into your Hugging Face cache folder when you first time generate something How to Get Installation Logs and How to Change Hugging Face Cache Folder : https://www.patreon.com/posts/108419878 Please make a fresh install When you enable all 4 optimizations the VRAM usages are like below Make sure shared VRAM is enabled because initial loading of the model need more VRAM Enable VAE Tiling + Enable VAE Slicing + Enable Model CPU Offload + Enable Sequential CPU Offload 1K (1024x1024) : 4 GB GPUs 2K (2048x2048) : 6 GB GPUs 4K (4096x4096) : 8 GB GPUs Still in any case may work on your GPU test it Just Enable VAE Tiling + Enable Model CPU Offload works great in many cases All below attached images are generated via SANA 4K model, they are RAW and their resolution is 5376x3072 Official repo page : https://github.com/NVlabs/Sana

reacted to MonsterMMORPG's post with ❤️ about 6 hours ago

It is now possible to generate 16 Megapixel (4096x4096) raw images with SANA 4K model using under 8GB VRAM, 4 Megapixel (2048x2048) images using under 6GB VRAM, and 1 Megapixel (1024x1024) images using under 4GB VRAM thanks to new optimizations 13 January 2024 Update Installers : https://www.patreon.com/posts/from-nvidia-labs-116474081 New 4K Tutorial Video : https://youtu.be/GjENQfHF4W8 Now the APP will use Diffusers Pipeline and it has huge VRAM optimizations You need to reinstall The models will be downloaded into your Hugging Face cache folder when you first time generate something How to Get Installation Logs and How to Change Hugging Face Cache Folder : https://www.patreon.com/posts/108419878 Please make a fresh install When you enable all 4 optimizations the VRAM usages are like below Make sure shared VRAM is enabled because initial loading of the model need more VRAM Enable VAE Tiling + Enable VAE Slicing + Enable Model CPU Offload + Enable Sequential CPU Offload 1K (1024x1024) : 4 GB GPUs 2K (2048x2048) : 6 GB GPUs 4K (4096x4096) : 8 GB GPUs Still in any case may work on your GPU test it Just Enable VAE Tiling + Enable Model CPU Offload works great in many cases All below attached images are generated via SANA 4K model, they are RAW and their resolution is 5376x3072 Official repo page : https://github.com/NVlabs/Sana

reacted to MonsterMMORPG's post with 🚀 about 6 hours ago

It is now possible to generate 16 Megapixel (4096x4096) raw images with SANA 4K model using under 8GB VRAM, 4 Megapixel (2048x2048) images using under 6GB VRAM, and 1 Megapixel (1024x1024) images using under 4GB VRAM thanks to new optimizations 13 January 2024 Update Installers : https://www.patreon.com/posts/from-nvidia-labs-116474081 New 4K Tutorial Video : https://youtu.be/GjENQfHF4W8 Now the APP will use Diffusers Pipeline and it has huge VRAM optimizations You need to reinstall The models will be downloaded into your Hugging Face cache folder when you first time generate something How to Get Installation Logs and How to Change Hugging Face Cache Folder : https://www.patreon.com/posts/108419878 Please make a fresh install When you enable all 4 optimizations the VRAM usages are like below Make sure shared VRAM is enabled because initial loading of the model need more VRAM Enable VAE Tiling + Enable VAE Slicing + Enable Model CPU Offload + Enable Sequential CPU Offload 1K (1024x1024) : 4 GB GPUs 2K (2048x2048) : 6 GB GPUs 4K (4096x4096) : 8 GB GPUs Still in any case may work on your GPU test it Just Enable VAE Tiling + Enable Model CPU Offload works great in many cases All below attached images are generated via SANA 4K model, they are RAW and their resolution is 5376x3072 Official repo page : https://github.com/NVlabs/Sana

View all activity

Organizations

AtAndDev's activity

upvoted a paper 9 days ago

2 OLMo 2 Furious

Paper • 2501.00656 • Published 13 days ago • 15

upvoted a paper 11 days ago

1.58-bit FLUX

Paper • 2412.18653 • Published 20 days ago • 69

upvoted a paper 23 days ago

Instruction Pre-Training: Language Models are Supervised Multitask Learners

Paper • 2406.14491 • Published Jun 20, 2024 • 87

upvoted a paper 29 days ago

Phi-4 Technical Report

Paper • 2412.08905 • Published Dec 12, 2024 • 101

upvoted a paper 30 days ago

The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale

Paper • 2406.17557 • Published Jun 25, 2024 • 90

upvoted a paper about 1 month ago

KTO: Model Alignment as Prospect Theoretic Optimization

Paper • 2402.01306 • Published Feb 2, 2024 • 16

upvoted a collection about 1 month ago

NuminaMath

Datasets and models for training SOTA math LLMs. See our GitHub for training & inference code: https://github.com/project-numina/aimo-progress-prize • 6 items • Updated Jul 21, 2024 • 70

upvoted 2 papers about 1 month ago

Large Language Models Can Self-Improve in Long-context Reasoning

Paper • 2411.08147 • Published Nov 12, 2024 • 63

O1 Replication Journey -- Part 2: Surpassing O1-preview through Simple Distillation, Big Progress or Bitter Lesson?

Paper • 2411.16489 • Published Nov 25, 2024 • 41

upvoted 2 collections about 2 months ago

🧠 Reasoning Models

7 items • Updated 14 days ago • 36

QwQ

Qwen with Questions • 2 items • Updated Nov 28, 2024 • 56

upvoted 4 papers about 2 months ago

Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking

Paper • 2403.09629 • Published Mar 14, 2024 • 75

V-STaR: Training Verifiers for Self-Taught Reasoners

Paper • 2402.06457 • Published Feb 9, 2024 • 9

Let's Verify Step by Step

Paper • 2305.20050 • Published May 31, 2023 • 10

Learn Beyond The Answer: Training Language Models with Reflection for Mathematical Reasoning

Paper • 2406.12050 • Published Jun 17, 2024 • 19

upvoted a collection about 2 months ago

Top LLM

Collection of TOP Open Source LLM, Sort by Best on top • 6 items • Updated Jul 26, 2024 • 13

upvoted 4 papers 2 months ago

Large Language Models Orchestrating Structured Reasoning Achieve Kaggle Grandmaster Level

Paper • 2411.03562 • Published Nov 5, 2024 • 65

What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspective

Paper • 2410.23743 • Published Oct 31, 2024 • 59

LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinations

Paper • 2410.02707 • Published Oct 3, 2024 • 48

GPT-4o System Card

Paper • 2410.21276 • Published Oct 25, 2024 • 83