Hugging Face – Posts

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

All HF Hub posts

jasoncorkill

posted an update about 20 hours ago

Post

2252

Do you remember https://thispersondoesnotexist.com/ ? It was one of the first cases where the future of generative media really hit us. Humans are incredibly good at recognizing and analyzing faces, so they are a very good litmus test for any generative image model.

But none of the current benchmarks measure the ability of models to generate humans independently. So we built our own. We measure the models ability to generate a diverse set of human faces and using over 20'000 human annotations we ranked all of the major models on their ability to generate faces. Find the full ranking here:
https://app.rapidata.ai/mri/benchmarks/68af24ae74482280b62f7596

We have release the full underlying data publicly here on huggingface: Rapidata/Face_Generation_Benchmark

mitkox

posted an update 2 days ago

Post

3472

I just threw Qwen3-0.6B in BF16 into an on device AI drag race on AMD Strix Halo with vLLM:

564 tokens/sec on short 100-token sprints
96 tokens/sec on 8K-token marathons

TL;DR You don't just run AI on AMD. You negotiate with it.

The hardware absolutely delivers. Spoiler alert; there is exactly ONE configuration where vLLM + ROCm + Triton + PyTorch + Drivers + Ubuntu Kernel to work at the same time. Finding it required the patience of a saint

Consumer AMD for AI inference is the ultimate "budget warrior" play, insane performance-per-euro, but you need hardcore technical skills that would make a senior sysadmin nod in quiet respect.

1 reply

Kseniase

posted an update 2 days ago

Post

3408

7+ Main precision formats used in AI:

Precision is very important in AI as it shapes how accurate and efficient models are. It controls how finely numbers are represented, approximating real-world values with formats like fixed-point and floating-point. A recent BF16 → FP16 study renewed attention to precision impact.
Here are the main precision types used in AI, from full precision for training to ultra-low precision for inference:

1. FP32 (Float32):
Standard full-precision float used in most training: 1 sign bit, 8 exponent bits, 23 mantissa bits. Default for backward-compatible training and baseline numerical stability

2. FP16 (Float16) → https://arxiv.org/abs/2305.10947v6
Half-precision float. It balances accuracy and efficiency. 1 sign bit, 5 exponent bits, 10 mantissa bits. Common on NVIDIA Tensor Cores and mixed-precision setups. There’s now a new wave of using it in reinforcement learning: https://www.turingpost.com/p/fp16

3. BF16 (BFloat16) → https://cloud.google.com/blog/products/ai-machine-learning/bfloat16-the-secret-to-high-performance-on-cloud-tpus
Same dynamic range as FP32 but fewer mantissa bits: 1 sign bit, 8 exponent bits (same as FP32), 7 mantissa bits. It was developed by the research group Google Brain as part of their AI/ML infrastructure work at Google. Preferred on TPUs and modern GPUs

4. FP8 (E4M3 / E5M2) → https://proceedings.neurips.cc/paper_files/paper/2018/file/335d3d1cd7ef05ec77714a215134914c-Paper.pdf
Emerging standard for training and inference on NVIDIA Hopper (H100) and Blackwell (B200) tensor cores and AMD MI300. Also supported in NVIDIA’s Transformer Engine: https://developer.nvidia.com/blog/floating-point-8-an-introduction-to-efficient-lower-precision-ai-training/
E4M3 = 4 exponent, 3 mantissa bits
E5M2 = 5 exponent, 2 mantissa bits

Read further below ⬇️
If you like this, also subscribe to the Turing post: https://www.turingpost.com/subscribe

1 reply

danielhanchen

posted an update 3 days ago

Post

3141

You can now run Kimi K2 Thinking locally with our Dynamic 1-bit GGUFs: unsloth/Kimi-K2-Thinking-GGUF

We shrank the 1T model to 245GB (-62%) & retained ~85% of accuracy on Aider Polyglot. Run on >247GB RAM for fast inference.

We also collaborated with the Moonshot AI Kimi team on a system prompt fix! 🥰

Guide + fix details: https://docs.unsloth.ai/models/kimi-k2-thinking-how-to-run-locally

prithivMLmods

posted an update 3 days ago

Post

2592

Introducing Photo-Mate-v2, based on FLUX.1-Kontext-dev, for advanced image manipulation tasks. It supports transforming scenes into top-down/bottom-up perspectives, CAM-right/left-view and its reverse, as well as general kontext-specified object removal. Below is the list of demos and adapters.🔥🤗

➤ Spaces [Demo] : prithivMLmods/Kontext-Photo-Mate-v2

Kontext-Adapters :
✦ Kontext-Bottom-Up-View: prithivMLmods/Kontext-Bottom-Up-View
✦ Kontext-CAM-Right-View: prithivMLmods/Kontext-CAM-Right-View
✦ Kontext-Top-Down-View: prithivMLmods/Kontext-Top-Down-View
✦ Kontext-CAM-Left-View: prithivMLmods/Kontext-CAM-Left-View
✦ Kontext-CAM-Right-View: prithivMLmods/Kontext-CAM-Right-View
✦ Kontext-Unblur-Upscale: prithivMLmods/Kontext-Unblur-Upscale
✦ Kontext-0811-exp: prithivMLmods/Kontext-0811-exp

Photo-Mate Collection:
✦ Kontext CAM Angles: https://huggingface.co/collections/prithivMLmods/kontext-cam-angles
✦ i2i - Kontext (exp): https://huggingface.co/collections/prithivMLmods/i2i-kontext-exp
✦ LZO-1 (Lossless Zoom Operator): https://huggingface.co/collections/prithivMLmods/lzo-1-lossless-zoom-operator

Related-Apps:
✦ Photo-Mate [Version 1.0]: prithivMLmods/Photo-Mate-i2i
✦ Image Generation Apps [Collection]: https://huggingface.co/collections/prithivMLmods/image-generation-apps-collection

To know more about it, visit the app page or the respective model page!
@prithivMLmods

branikita

posted an update 3 days ago

Post

2836

In November, with

Robonine are starting work on the next version of the SO ARM 102 manipulator. The version will be open source and agreed upon with @therobotbuilder the creator of the original manipulator.

We are planning to:

- increase positioning accuracy by approximately 2x using Feetech STS 3250 motors
- increase working payload from 200g to 300g
- increase rigidity using parametric design optimization and stiffer plastic
- increase length to 550 mm
- increase folding angles
- use ISO 9409-1-50-4-M6 mounting standard for the gripper
- use a parallel gripper in the default version
- update the mounting plate for different camera types, M3 grid with 12.5 mm pitch
- add table mounting standard 80x80 M8

The number of degrees of freedom and basic kinematics will remain the same.

Are there other things missing for working with SO ARM 100?

- Any standard inputs/outputs, for example?
- Status indicators?
- Perhaps some types of mounting for third-party grippers are more preferable?
- Anything else?

codelion

posted an update 1 day ago

Post

2734

Want to experiment with pre-training dataset mixtures but don't want to process terabytes of data? We've got you covered.

We're releasing a collection of several carefully curated 1B token dataset samples specifically designed for rapid prototyping and pretraining experiments: https://huggingface.co/collections/codelion/pre-training-dataset-samples

These samples were created using reservoir sampling - an algorithm that guarantees statistically unbiased random samples from massive source datasets. This means results you get at the 1B token scale are representative of how these datasets behave at 100B+ token scales, letting you iterate quickly without the computational overhead.

The collection includes:
- finePDFs-1B: High-quality textbook-style educational content
- DCLM-baseline-1B: Filtered, diverse web content
- FineWeb-Edu-1B: Curated educational web resources

We used these exact samples to run 50+ systematic experiments on dataset mixing strategies, ultimately discovering that a 50-30-20 mixture of finePDFs + DCLM-baseline + FineWeb-Edu achieves 90%+ of GPT-2's performance with just 1/10th the training data.

Whether you're researching optimal data mixtures, testing curriculum learning strategies, or just want to quickly prototype a pretraining run, these samples give you a solid foundation to start experimenting immediately.

Read the full story of how we used these datasets to find the optimal pretraining recipe: https://huggingface.co/blog/codelion/optimal-dataset-mixing

1 reply

MonsterMMORPG

posted an update 3 days ago

Post

1819

Qwen Image Base Model Training vs FLUX SRPO Training 20 images comparison (top ones Qwen bottom ones FLUX) - Same Dataset (28 imgs) - I can't return back to FLUX such as massive difference - Qwen destroys the FLUX at complex prompts and emotions

Full tutorial link > https://www.youtube.com/watch?v=DPX3eBTuO_Y

Info
This is a full comprehensive step-by-step tutorial for how to train Qwen Image models. This tutorial covers how to do LoRA training and full Fine-Tuning / DreamBooth training on Qwen Image models. It covers both the Qwen Image base model and the Qwen Image Edit Plus 2509 model. This tutorial is the product of 21 days of full R&D, costing over $800 in cloud services to find the best configurations for training. Furthermore, we have developed an amazing, ultra-easy-to-use Gradio app to use the legendary Kohya Musubi Tuner trainer with ease. You will be able to train locally on your Windows computer with GPUs with as little as 6 GB of VRAM for both LoRA and Fine-Tuning. Furthermore, I have shown how to train a character (person), a product (perfume) and a style (GTA5 artworks).

Resources
The post used in tutorial to download zip file : https://www.patreon.com/posts/qwen-trainer-app-137551634

Requirements tutorial : https://youtu.be/DrhUHnYfwC0

SwarmUI tutorial : https://youtu.be/c3gEoAyL2IE

Used Prompts
https://gist.github.com/FurkanGozukara/069523015d18a3e63d74c59257447f5b

Comparison Images Full Size

2 replies

nouamanetazi

posted an update 12 days ago

Post

3592

After training 𝐒𝐦𝐨𝐥𝐋𝐌𝟑 on 𝟑𝟖𝟒 𝐇𝟏𝟎𝟎𝐬 for nearly a month, I've come to realize something most people overlook: 𝐢𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 𝐢𝐬 𝐭𝐡𝐞 𝐦𝐚𝐤𝐞-𝐨𝐫-𝐛𝐫𝐞𝐚𝐤 𝐟𝐚𝐜𝐭𝐨𝐫 𝐢𝐧 𝐋𝐋𝐌 𝐭𝐫𝐚𝐢𝐧𝐢𝐧𝐠. 🔥

Everyone talks about model architecture and data quality. And yes, those matter immensely. But here's what nobody tells you: when your training run fails at 2 AM because of mysterious 𝐍𝐂𝐂𝐋 𝐞𝐫𝐫𝐨𝐫𝐬, or when your expensive GPU cluster is running at 𝟔𝟎% 𝐞𝐟𝐟𝐢𝐜𝐢𝐞𝐧𝐜𝐲, the problem isn't your model. It's most probably a 𝐦𝐢𝐬𝐮𝐬𝐞 𝐨𝐟 𝐭𝐡𝐞 𝐡𝐚𝐫𝐝𝐰𝐚𝐫𝐞. 🛠️

Questions that seemed simple but had no clear answers: Why is 𝐌𝐨𝐄 𝐭𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐬𝐥𝐨𝐰𝐞𝐫 𝐭𝐡𝐚𝐧 𝐝𝐞𝐧𝐬𝐞 𝐦𝐨𝐝𝐞𝐥𝐬? Which 𝐍𝐂𝐂𝐋 𝐟𝐥𝐚𝐠𝐬 should we actually set? How often should we checkpoint without killing throughput?

That's why we built 𝐓𝐡𝐞 𝐒𝐦𝐨𝐥 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐏𝐥𝐚𝐲𝐛𝐨𝐨𝐤 📖: a complete guide covering everything from model architecture and data curation to the SmolLM3 training marathon, post-training techniques, and crucially, the 𝐢𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 𝐥𝐚𝐲𝐞𝐫 that most teams get wrong.

We validated real vs theoretical bandwidth across the entire stack: 𝐇𝐁𝐌𝟑 𝐡𝐢𝐭𝐭𝐢𝐧𝐠 𝟑 𝐓𝐁/𝐬, 𝐍𝐕𝐋𝐢𝐧𝐤 𝟒.𝟎 𝐫𝐞𝐚𝐜𝐡𝐢𝐧𝐠 𝟕𝟖𝟔 𝐆𝐁/𝐬, 𝐏𝐂𝐈𝐞 𝐆𝐞𝐧𝟒 𝐚𝐭 𝟏𝟒.𝟐 𝐆𝐁/𝐬. Then we ran collective operations across 𝟏𝟐𝟖 𝐆𝐏𝐔𝐬 (16 nodes, 8xH100s each) and measured how performance degrades at scale: all-reduce drops from 𝟒𝟖𝟎 𝐆𝐁/𝐬 on a single node to 𝟑𝟐𝟎-𝟑𝟓𝟎 𝐆𝐁/𝐬 across 16 nodes.

If you've ever wondered why your training runs are slower than they should be, or you're planning to scale up and want to avoid expensive mistakes, this guide might save you weeks of debugging.

𝐓𝐡𝐞 𝐒𝐦𝐨𝐥 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐏𝐥𝐚𝐲𝐛𝐨𝐨𝐤: https://lnkd.in/e5MKXUHS

Shared with ❤️ by the HuggingFace team

wang12390

posted an update 1 day ago

Post

654

Garment Try-On: Transforming Fashion Design and Shopping with AI

Discover how AI-powered Garment Try-On technology transforms fashion design, e-commerce, and sustainability with realistic 3D garment visualization.

Miragic-AI/Miragic-Virtual-Try-On

Recently active users