🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

New embedding mixture-of-experts model by Nomic AI

We will release the quantized models of Qwen3 to you in the following days. Today we release the AWQ and GGUFs of Qwen3-14B and Qwen3-32B, which enables using the models with limited GPU memory. Qwen3-32B-AWQ: huggingface.co/Qwen/Qwen3-32B… Qwen3-32B-GGUF: huggingface.co/Qwen/Qwen3-32B…

llama.cpp shipped new support for vision models this morning, including macOS binaries (albeit quarantined so you have to take extra steps to run them) that let you run vision models in a terminal or as a localhost web UI

Son has been doing an outstanding job at maintaining the llama-server implementation and now bringing full-blown vision input support to llama.cpp! Massive kudos and thanks for your valuable contributions to the project!

llama.cpp is now fully compatible with VLMs 💥 HUGE kudos to Xuan-Son Nguyen from HF and to the ggml team 💟 Here are a selection of pre-quantized models, ready to be used, from: - Google DeepMind Gemma - Mistral AI Pixtral - Qwen VL - Hugging Face SmolVLM Give them a

Llama.cpp now supports vision models thanks to Xuan-Son Nguyen Georgi Gerganov! On-device multi-modal🔥🔥🔥

Real-time webcam demo with Hugging Face SmolVLM and ggml llama.cpp server. All running locally on a Macbook M3

👀

Deploy vision models with llama.cpp on Hugging Face

PSA for applications that use local AI models - here is how to do it right: More and more applications are adding support for local AI models, which is great. But I notice that they are doing it the wrong way (see the screenshots below). The right way to do it is to add a

Summarize latest Fireship's video using cutting edge ggml's llama.cpp audio support (model: Ultravox + Llama 3.1 8B)

llama.cpp streaming support for tool calling & thoughts was just merged: please test & report any issues 😅 github.com/ggml-org/llama… #llamacpp

llm-llama-server now supports tools, which means this local Gemma demo should work (if you have 3.2GB free): brew install llama.cpp llama-server --jinja -hf unsloth/gemma-3-4b-it-GGUF:Q4_K_XL uvx --with llm-llama-server llm -m llama-server-tools -T llm_time 'what time is it?'

🎙️ After serving millions of users through our text-to-speech platform, one need kept coming up: fine-grained AI speech editing - the ability to modify existing speech. Today, we’re open-sourcing PlayDiffusion, a diffusion-based inpainting model built for that exact purpose.

Diffusion in-painting and TTS model. Give it a try! We are expecting you to build some great things with it. Let me know if you need any help.

Massive QoL update: You can now filter the models on the hub based on their size 🔥 Find the model that fits YOUR needs faster ⚡️

👀 curious if this would fly