nyuuzyou (nyuuzyou)

reacted to codelion's post with 🔥 5 days ago

Post

2298

Introducing AutoThink: Adaptive reasoning for LLMs that improves performance by 43% on reasoning benchmarks!

Instead of using fixed thinking budgets, AutoThink:
- Classifies query complexity (HIGH/LOW) using adaptive classification
- Dynamically allocates thinking tokens based on complexity
- Uses steering vectors derived from Pivotal Token Search to guide reasoning patterns

Results on DeepSeek-R1-Distill-Qwen-1.5B:
- GPQA-Diamond: 31.06% vs 21.72% baseline (+9.34 points)
- MMLU-Pro: 26.38% vs 25.58% baseline (+0.8 points)
- Uses fewer tokens than baseline approaches

Works with any local reasoning model - DeepSeek, Qwen, Llama, custom models. The technique combines our research on Pivotal Token Search (PTS) implementation and adaptive classification frameworks.

Paper: AutoThink: efficient inference for reasoning LLMs
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5253327

Code and examples:
https://github.com/codelion/optillm/tree/main/optillm/autothink

PTS implementation and technical details:
https://github.com/codelion/pts
https://huggingface.co/blog/codelion/pts

Adaptive classifier framework:
https://github.com/codelion/adaptive-classifier

Would love to hear your thoughts on adaptive resource allocation for LLM reasoning! Have you experimented with similar approaches?

5 replies

·

reacted to ProCreations's post with 🚀 6 days ago

Post

2852

Eyyyy 50 followers 🤯

1 reply

·

replied to jasoncorkill's post 7 days ago

I'm not sure if this is it, but it looks like it's already available

https://huggingface.co/datasets/Rapidata/text-2-video-human-preferences-veo3
https://huggingface.co/datasets/Rapidata/1k-ranked-videos-coherence

posted an update 10 days ago

Post

2900

I recently updated nyuuzyou/pxhere dataset and it now contains approximately 1.1M CC0 high-resolution images

reacted to KaraKaraWitch's post with 🔥 10 days ago

Post

2616

> New Model
> Looks at Model Card
> "Open-Weights"

1 reply

·

replied to merve's post 11 days ago

ये जापानी भाषा को सपोर्ट नहीं करता।

facebook/nllb-200-distilled-600M supports Japanese, maybe it's worth a bug report?

posted an update 16 days ago

Post

571

🎬 Animations Dataset - nyuuzyou/animations

Collection of approximately 50,849 animations featuring:

- Simple sketch-level drawings, text elements, and embedded images
- Includes metadata: unique UUID, frame components, timing between frames, and canvas settings
- Contains full-resolution frames and source images in original format
- Organized with approximately 1,000 animations per archive

All animations are stored in a structured format that preserves relationships between component images and their composition within animation frames.

reacted to cbensimon's post with 🔥 17 days ago

Post

5711

🚀 ZeroGPU medium size is now available as a power-user feature

Nothing too fancy for now—ZeroGPU Spaces still default to large (70GB VRAM)—but this paves the way for:
- 💰 size-based quotas / pricing (medium will offer significantly more usage than large)
- 🦣 the upcoming xlarge size (141GB VRAM)

You can as of now control GPU size via a Space variable. Accepted values:
- auto (future default)
- medium
- large (current default)

The auto mode checks total CUDA tensor size during startup:
- More than 30GB → large
- Otherwise → medium

3 replies

·

reacted to seawolf2357's post with 👀 17 days ago

Post

6024

Samsung Hacking Incident: Samsung Electronics' Official Hugging Face Account Compromised
Samsung Electronics' official Hugging Face account has been hacked. Approximately 17 hours ago, two new language models (LLMs) were registered under Samsung Electronics' official Hugging Face account. These models are:

https://huggingface.co/Samsung/MuTokenZero2-32B
https://huggingface.co/Samsung/MythoMax-L2-13B

The model descriptions contain absurd and false claims, such as being trained on "1 million W200 GPUs," hardware that doesn't even exist.
Moreover, community participants on Hugging Face who have noticed this issue are continuously posting that Samsung Electronics' account has been compromised.
There is concern about potential secondary and tertiary damage if users download these LLMs released under the Samsung Electronics account, trusting Samsung's reputation without knowing about the hack.
Samsung Electronics appears to be unaware of this situation, as they have not taken any visible measures yet, such as changing the account password.
Source: https://discord.gg/openfreeai

2 replies

·

reacted to fdaudens's post with ❤️ 19 days ago

Post

5102

Tried something new: an AI-generated podcast that breaks down the top research paper each day. Fully automated, now live on Spotify.

I built this prototype to help keep up with the rapid pace of AI developments and, hopefully, make cutting-edge research more accessible. I don’t know about you, but just listening to a conversation about a paper really helps the content sink in for me.

This build taught me a lot about full automation. If you’re into the technical weeds: Qwen3 runs on Inference to handle the script, Kokoro does the voice, and the whole thing gets published automatically thanks to the Hugging Face Jobs API and Gradio deployment.

It’s not perfect yet — I’ll be monitoring for hallucinations and incoherence. The voice model still needs polish, but it’s a promising start. Would love to build this with the community — submit a PR or send feedback. It’s just a beta of an experimental idea!

Big kudos to @m-ric , whose Open NotebookLM this is based on, and to @nielsr for his terrific work making research papers more accessible.

- Podcast on Spotify: https://open.spotify.com/show/3PTucIW1w1GIkqTYm32ka7?si=c7a851f83e6d4331 (Apple Podcasts soon)
- Code: fdaudens/podcast-jobs
- Open NotebookLM: m-ric/open-notebooklm
- Also super helpful, @qgallouedec 's tutorial on HF Jobs API: qgallouedec/run-hello-world

1 reply

·

posted an update 19 days ago

Post

568

🖼️ PxHere Images Dataset - nyuuzyou/pxhere

Collection of approximately 760,600 high-quality photographs featuring:

- Sourced from pxhere.com across diverse subjects and styles
- Includes metadata: unique image ID, tags, upload date, and detailed EXIF information
- Contains full-resolution images in original format (typically JPEG)
- Organized in WebDataset format with approximately 1,000 images per archive

All images are in the public domain, requiring no attribution for personal or commercial use.

reacted to dhruv3006's post with 🚀 21 days ago

Post

2403

The era of local Computer Use AI Agents is here.

Meet UI-TARS-1.5-7B-6bit, now running natively on Apple Silicon via MLX.

The video is of UI-TARS-1.5-7B-6bit completing the prompt "draw a line from the red circle to the green circle, then open reddit in a new tab" running entirely on MacBook. The video is just a replay, during actual usage it took between 15s to 50s per turn with 720p screenshots (on avg its ~30s per turn), this was also with many apps open so it had to fight for memory at times.

Built using c/ua : https://github.com/trycua/cua

Join us making them here: https://discord.gg/4fuebBsAUj

Kudos to the MLX community here on huggingface :

mlx-community

posted an update 25 days ago

Post

757

🎞️ HailuoAI Video Metadata Dataset - nyuuzyou/hailuoai

Collection of 544,646 AI-generated video metadata entries from HailuoAI featuring:

- Comprehensive metadata: direct video URLs, dimensions, creation parameters, model IDs, tags, and more.
- All metadata explicitly released into the public domain under the CC0 license.
- Organized in a single train split with 544,646 entries.

This is likely the most extensive public dataset of AI-generated videos to date.

reacted to m-ric's post with 🔥 25 days ago

Post

4411

I've made an open version of Google's NotebookLM, and it shows the superiority of the open source tech task! 💪

The app's workflow is simple. Given a source PDF or URL, it extracts the content from it, then tasks Meta's Llama 3.3-70B with writing the podcast script, with a good prompt crafted by @gabrielchua ("two hosts, with lively discussion, fun notes, insightful question etc.")
Then it hands off the text-to-speech conversion to Kokoro-82M, and there you go, you have two hosts discussion any article.

The generation is nearly instant, because:
> Llama 3.3 70B is running at 1,000 tokens/seconds with Cerebras inference
> The audio is generated in streaming mode by the tiny (yet powerful) Kokoro, generating voices faster than real-time.

And the audio generation runs for free on Zero GPUs, hosted by HF on H200s.

Overall, open source solutions rival the quality of closed-source solutions at close to no cost!

Try it here 👉👉 m-ric/open-notebooklm

3 replies

·

reacted to onekq's post with 🤗 25 days ago

Post

3279

This time Gemini is very quick with API support on its 2.5 pro May release. The performance is impressive too, now it is among top contenders like o4, R1, and Claude.

onekq-ai/WebApp1K-models-leaderboard

replied to clem's post 27 days ago

Typically, I handpick potentially suitable models for a given task using a small dataset. Finally, the model with the best speed or tokens per dollar is used among suitable variants.

posted an update 29 days ago

Post

3683

nyuuzyou/svgfind 👀

Well, everything happens for the first time 🤗. Thank you all!

reacted to ginipick's post with 🔥 about 1 month ago

Post

5202

🔮 Mistral Perflexity AI - Local LLM Space with Web Search Capabilities 🌐
Hello AI enthusiasts! Today I'm excited to introduce my special Hugging Face space! 🚀

ginigen/Mistral-Perflexity

✨ Key Features

Powerful Model: Using Private-BitSix-Mistral-Small-3.1-24B-Instruct-2503, optimized through 6-bit quantization to run smoothly on local 4090 GPUs! 💪
Web Search Integration: Leveraging the Brave Search API to provide real-time web search results for user queries! 🔍
Customizable Responses: Shape AI personality and response format through system messages ⚙️
Multilingual Support: Perfect handling of both English and Korean! 🇺🇸🇰🇷

🛠️ Technical Highlights

GGUF Format: Optimized quantized model with excellent memory efficiency
Flash Attention: Applied optimization technology for faster inference speeds
8K Context Window: Capable of handling lengthy conversations and complex queries
Streaming Responses: Watch text being generated in real-time

💡 Use Cases

Complex Q&A requiring real-time information
Programming assistance and code generation
Multilingual content creation and translation
Summarization and explanation of learning materials

🔧 Customization
Adjust various parameters like Temperature, Top-p, Top-k, and repetition penalty to control response creativity and accuracy. Lower temperature (0.1-0.5) produces more deterministic responses, while higher values (0.7-1.0) generate more creative outputs!

🌟 Try It Yourself!
This space is available for anyone to use for free. Experience the power of a robust local LLM combined with web search capabilities! Your feedback is always welcome! 😊

posted an update about 1 month ago

Post

3626

🖼️ PublicDomainFiles.com Collection - nyuuzyou/publicdomainfiles

Collection of 206,204 Public Domain multimedia files featuring:

- Comprehensive metadata: title, description, creator name, keywords, original page URL, and more.
- Contains various media types including images, clip art, artwork, fonts, videos, and TV shows.
- All content explicitly released into the public domain under the CC0 license.
- Organized in a single train split with 206,204 entries.

reacted to openfree's post with 🔥 about 1 month ago

Post

6087

🔥 Creating a qwen3-30b-a3b / qwen3-235b-a22b Chatbot with Deep Research Capabilities 🚀

openfree/qwen3-30b-a3b-research
openfree/qwen3-235b-a22b-research

Hello AI researchers! 👋 Today I'm introducing a powerful chatbot implementation with real-time web search capabilities.
✨ Key Features

🧠 Chatbot based on qwen3-30b-a3b and llama4-maverick models
🔍 LLM-based optimal keyword extraction
🌐 Real-time web search using SerpHouse API
💬 Streaming responses for natural conversation experience

🛠️ Technology Stack

Gradio: Implementation of intuitive web interface
Fireworks.ai API: Access to high-performance LLM models
SerpHouse API: Collection of real-time search results

🌟 Application Areas

Question answering systems requiring up-to-date information
Providing current information beyond training data
Delivering reliable information with accurate sources

Add real-time search capabilities to your AI applications with this project! 🎉 Leave your questions or suggestions in the comments! Let's improve it together~ 💪
#LLM #ArtificialIntelligence #WebSearch #Gradio #DeepResearch #OpenSource

nyuuzyou

AI & ML interests

Recent Activity

Organizations

nyuuzyou's activity