Hugging Face's logo
Joseph's picture
146 15 399

Joseph

Joseph717171
InferenceIllusionist's profile picture Erland's profile picture helloAI333's profile picture
·

AI & ML interests

None yet

Recent Activity

reacted to merterbak's post with 🤗 about 6 hours ago
Meta has unveiled its Llama 4 🦙 family of models, featuring native multimodality and mixture-of-experts architecture. Two model families are available now: Models🤗: https://huggingface.co/collections/meta-llama/llama-4-67f0c30d9fe03840bc9d0164 Blog Post: https://ai.meta.com/blog/llama-4-multimodal-intelligence/ HF's Blog Post: https://huggingface.co/blog/llama4-release - 🧠 Native Multimodality - Process text and images in a unified architecture - 🔍 Mixture-of-Experts - First Llama models using MoE for incredible efficiency - 📏 Super Long Context - Up to 10M tokens - 🌐 Multilingual Power - Trained on 200 languages with 10x more multilingual tokens than Llama 3 (including over 100 languages with over 1 billion tokens each) 🔹 Llama 4 Scout - 17B active parameters (109B total) - 16 experts architecture - 10M context window - Fits on a single H100 GPU - Beats Gemma 3, Gemini 2.0 Flash-Lite, and Mistral 3.1 🔹 Llama 4 Maverick - 17B active parameters (400B total) - 128 experts architecture - It can fit perfectly on DGX H100(8x H100) - 1M context window - Outperforms GPT-4o and Gemini 2.0 Flash - ELO score of 1417 on LMArena currently second best model on arena 🔹 Llama 4 Behemoth (Coming Soon) - 288B active parameters (2T total) - 16 experts architecture - Teacher model for Scout and Maverick - Outperforms GPT-4.5, Claude Sonnet 3.7, and Gemini 2.0 Pro on STEM benchmarks
reacted to merterbak's post with 🔥 about 6 hours ago
Meta has unveiled its Llama 4 🦙 family of models, featuring native multimodality and mixture-of-experts architecture. Two model families are available now: Models🤗: https://huggingface.co/collections/meta-llama/llama-4-67f0c30d9fe03840bc9d0164 Blog Post: https://ai.meta.com/blog/llama-4-multimodal-intelligence/ HF's Blog Post: https://huggingface.co/blog/llama4-release - 🧠 Native Multimodality - Process text and images in a unified architecture - 🔍 Mixture-of-Experts - First Llama models using MoE for incredible efficiency - 📏 Super Long Context - Up to 10M tokens - 🌐 Multilingual Power - Trained on 200 languages with 10x more multilingual tokens than Llama 3 (including over 100 languages with over 1 billion tokens each) 🔹 Llama 4 Scout - 17B active parameters (109B total) - 16 experts architecture - 10M context window - Fits on a single H100 GPU - Beats Gemma 3, Gemini 2.0 Flash-Lite, and Mistral 3.1 🔹 Llama 4 Maverick - 17B active parameters (400B total) - 128 experts architecture - It can fit perfectly on DGX H100(8x H100) - 1M context window - Outperforms GPT-4o and Gemini 2.0 Flash - ELO score of 1417 on LMArena currently second best model on arena 🔹 Llama 4 Behemoth (Coming Soon) - 288B active parameters (2T total) - 16 experts architecture - Teacher model for Scout and Maverick - Outperforms GPT-4.5, Claude Sonnet 3.7, and Gemini 2.0 Pro on STEM benchmarks
new activity about 13 hours ago
google/gemma-3-12b-it-qat-q4_0-gguf:Permission to access the Unquantized versions of the QAT weights for Gemma-3
View all activity

Organizations

Hugging Face Discord Community's profile picture

Collections 3

GGUF Llama-3.2-Instruct-OQ8_0-F32.EF32.IQ4_K-Q8_0 IQuants
Custom GGUF quants of Meta’s Llama-3.2-Instruct's finetunes, where the Output Tensors are quantized to Q8_0 or F32 and the Embeddings are kept @F32
GGUF LLaMa-3.1-8B-Instruct-Based OQ8_0.EF32 IQuants
Custom GGUF quants of Llama-3.1-8B-Instruct fine-tunes, where the Output Tensors are quantized to Q8_0 while the Embeddings are kept at F32. 🧠🔥🚀

datasets

None public yet