Running
2
π§Έ
Adam Molnar
lunarflu
AI & ML interests
join the Hugging Face discord! hf.co/discord/join
Recent Activity
liked
a Space
about 11 hours ago
khouraisan/fumo-classifier
upvoted
an
article
1 day ago
Fine-tune ModernBERT for text classification using synthetic data
Organizations
lunarflu's activity
liked
a
Space
about 11 hours ago
upvoted
4
articles
1 day ago
Article
π¦Έπ»#2: Your Go-To Vocabulary to Navigate the World of AI Agents and Agentic Workflows
By
β’
β’
8Article
Fine-tune ModernBERT for text classification using synthetic data
By
β’
β’
16Article
Debate Championship for LLMs
By
β’
β’
4Article
Finetuning Falcon 7b in a hybrid distributed fashion
By
β’
β’
3
upvoted
a
paper
1 day ago
waiting for moonshine-distilled next :)
reacted to
Xenova's
post with ππ₯β€οΈ
6 days ago
Post
2610
Introducing Moonshine Web: real-time speech recognition running 100% locally in your browser!
π Faster and more accurate than Whisper
π Privacy-focused (no data leaves your device)
β‘οΈ WebGPU accelerated (w/ WASM fallback)
π₯ Powered by ONNX Runtime Web and Transformers.js
Demo: webml-community/moonshine-web
Source code: https://github.com/huggingface/transformers.js-examples/tree/main/moonshine-web
π Faster and more accurate than Whisper
π Privacy-focused (no data leaves your device)
β‘οΈ WebGPU accelerated (w/ WASM fallback)
π₯ Powered by ONNX Runtime Web and Transformers.js
Demo: webml-community/moonshine-web
Source code: https://github.com/huggingface/transformers.js-examples/tree/main/moonshine-web
-
- 5 replies
reacted to
ginipick's
post with π₯
8 days ago
Post
4440
π¬ Revolutionize Your Video Creation
Dokdo Multimodal AI Transform a single image into a stunning video with perfect audio harmony! π
Superior Technology π«
Advanced Flow Matching: Smoother video transitions surpassing Kling and Sora
Intelligent Sound System: Automatically generates perfect audio by analyzing video mood
Multimodal Framework: Advanced AI integrating image, text, and audio analysis
Outstanding Performance π―
Ultra-High Resolution: 4K video quality with bfloat16 acceleration
Real-Time Optimization: 3x faster processing with PyTorch GPU acceleration
Smart Sound Matching: Real-time audio effects based on scene transitions and motion
Exceptional Features β¨
Custom Audio Creation: Natural soundtrack matching video tempo and rhythm
Intelligent Watermarking: Adaptive watermark adjusting to video characteristics
Multilingual Support: Precise translation engine powered by Helsinki-NLP
Versatile Applications π
Social Media Marketing: Create engaging shorts for Instagram and YouTube
Product Promotion: Dynamic promotional videos highlighting product features
Educational Content: Interactive learning materials with enhanced engagement
Portfolio Enhancement: Professional-grade videos showcasing your work
Experience the video revolution with Dokdo Multimodal, where anyone can create professional-quality content from a single image. Elevate your content with perfectly synchronized video and audio that captivates your audience! π¨
Start creating stunning videos that stand out from the crowd - whether you're a marketer, educator, content creator, or business owner. Join the future of AI-powered video creation today!
ginipick/Dokdo-multimodal
#VideoInnovation #AITechnology #PremiumContent #MarketingSolution
π Please turn on your sound for the best viewing experience!
Dokdo Multimodal AI Transform a single image into a stunning video with perfect audio harmony! π
Superior Technology π«
Advanced Flow Matching: Smoother video transitions surpassing Kling and Sora
Intelligent Sound System: Automatically generates perfect audio by analyzing video mood
Multimodal Framework: Advanced AI integrating image, text, and audio analysis
Outstanding Performance π―
Ultra-High Resolution: 4K video quality with bfloat16 acceleration
Real-Time Optimization: 3x faster processing with PyTorch GPU acceleration
Smart Sound Matching: Real-time audio effects based on scene transitions and motion
Exceptional Features β¨
Custom Audio Creation: Natural soundtrack matching video tempo and rhythm
Intelligent Watermarking: Adaptive watermark adjusting to video characteristics
Multilingual Support: Precise translation engine powered by Helsinki-NLP
Versatile Applications π
Social Media Marketing: Create engaging shorts for Instagram and YouTube
Product Promotion: Dynamic promotional videos highlighting product features
Educational Content: Interactive learning materials with enhanced engagement
Portfolio Enhancement: Professional-grade videos showcasing your work
Experience the video revolution with Dokdo Multimodal, where anyone can create professional-quality content from a single image. Elevate your content with perfectly synchronized video and audio that captivates your audience! π¨
Start creating stunning videos that stand out from the crowd - whether you're a marketer, educator, content creator, or business owner. Join the future of AI-powered video creation today!
ginipick/Dokdo-multimodal
#VideoInnovation #AITechnology #PremiumContent #MarketingSolution
π Please turn on your sound for the best viewing experience!
- 1 reply
reacted to
vincentg64's
post with π₯
8 days ago
Post
2186
LLM 2.0, RAG & Non-Standard Gen AI on GitHub https://mltblog.com/3DsyZSq
In this article, I share my latest Gen AI and LLM advances, featuring innovative approaches radically different from both standard AI and classical ML/NLP. The focus is on doing better with less, using efficient architectures, new algorithms and evaluation metrics. It originates from research that I started long ago. It gained significant momentum in the last two years. See background and history at https://mltblog.com/4g2sKTv.
OpenAI, Perplexity, Anthropic, Llama and others typically follow the trend and implement solutions very similar to mines within 3 to 6 months after I publish new milestones. For instance, multi-tokens, knowledge graph tokens, multi-indexes, real-time fine-tuning, mixtures of experts, LLM routers, small enterprise sub-LLMs, prompt distillation, relevancy scoring engine, deep contextual retrieval, optimum agentic chunking, and modern UI instead of the basic prompt box. I keep adding new features all the time, staying ahead of competition.
β‘οΈ Read full article with links to GitHub, at https://mltblog.com/3DsyZSq
In this article, I share my latest Gen AI and LLM advances, featuring innovative approaches radically different from both standard AI and classical ML/NLP. The focus is on doing better with less, using efficient architectures, new algorithms and evaluation metrics. It originates from research that I started long ago. It gained significant momentum in the last two years. See background and history at https://mltblog.com/4g2sKTv.
OpenAI, Perplexity, Anthropic, Llama and others typically follow the trend and implement solutions very similar to mines within 3 to 6 months after I publish new milestones. For instance, multi-tokens, knowledge graph tokens, multi-indexes, real-time fine-tuning, mixtures of experts, LLM routers, small enterprise sub-LLMs, prompt distillation, relevancy scoring engine, deep contextual retrieval, optimum agentic chunking, and modern UI instead of the basic prompt box. I keep adding new features all the time, staying ahead of competition.
β‘οΈ Read full article with links to GitHub, at https://mltblog.com/3DsyZSq
- 1 reply
updated
2
Spaces
8 days ago
liked
a
model
9 days ago
liked
a
Space
9 days ago
reacted to
merve's
post with ππβ€οΈ
9 days ago
Post
2722
Aya by Cohere For AI can now see! π
C4AI community has built Maya 8B, a new open-source multilingual VLM built on SigLIP and Aya 8B π± works on 8 languages! π£οΈ
The authors extend Llava dataset using Aya's translation capabilities with 558k examples!
ry it here kkr5155/maya_demo
Dataset maya-multimodal/pretrain
Model maya-multimodal/maya π
kudos @nahidalam and team
C4AI community has built Maya 8B, a new open-source multilingual VLM built on SigLIP and Aya 8B π± works on 8 languages! π£οΈ
The authors extend Llava dataset using Aya's translation capabilities with 558k examples!
ry it here kkr5155/maya_demo
Dataset maya-multimodal/pretrain
Model maya-multimodal/maya π
kudos @nahidalam and team
- 1 reply