Hugging Models
@HuggingModels
We're sharing/showcasing best of @huggingface models. Follow to stay in loop. Promoting Open-Source models.
33 Following 72.3K Followers
🔗 Explore the full model details here:
With 421 downloads and 61 likes, this FP4 variant proves small can be mighty. Get near full-precision quality at a fraction of the cost. Ideal for edge devices, prototyping, and budget-friendly AI experiments.
Show more
Built on the glm5_next architecture and optimized with ModelOpt, this model uses FP4 quantization to shrink weights dramatically. Safetensors format ensures safe, fast loading. It's a lean take on a powerful multimodal design.
Show more
Feed it images and text, get text back. Build visual Q&A bots, image captioning tools, or document analyzers. The FP4 format means you can run it on consumer hardware without sacrificing GLM-5's reasoning.
Show more
Meet GLM-5.3-Flash-FP4-Local: a quantized image-text-to-text model from NVIDIA. FP4 precision slashes memory use while keeping GLM-5's multimodal smarts. Perfect for local deployment on modest GPUs.
Show more
🔗 Explore the full model details here:
53K+ downloads and growing. Why? It delivers vision-language tasks at a fraction of the size. Ternary weights mean faster loads, lower memory, and broad hardware support. Big capability in a bonsai-sized package.
Show more
Built on a 9B parameter architecture with hybrid attention for efficiency. The ternary 2-bit quantization keeps it tiny while preserving multimodal understanding. GGUF format means it plays nice with llama.cpp across CUDA and Metal backends.
Show more
Point it at a photo and ask questions, generate captions, or extract details. Perfect for on-device vision assistants, accessibility tools, or smart photo search. No cloud needed, just local inference with llama.cpp.
Show more
Meet Ternary-Bonsai-2-9B-GGUF: a 9B image-text-to-text model squeezed into 2-bit ternary weights. Runs on llama.cpp with CUDA and Metal, so you can chat about images right on your laptop or phone. Small but mighty!
Show more
Meet Ternary-Bonsai-2-27B: a 27B parameter model squeezed into 2-bit ternary format. Runs on llama.cpp with CUDA and Metal support. 405K downloads and counting. On-device AI just got a whole lot lighter.
Show more
Meet Qwen3.8-27B-OBLITERATED-Mythos-Class-Agentic. It's a 27B text-gen model that's abliterated for uncensored outputs, plus it's built for agentic tool calling. Think of it as a creative powerhouse with a rebellious streak. #AI# #LLM#
Show more
🚨Just in: Apple maybe behind mystery model Ox Alpha.
Ever wanted an AI that can 'see' medical images and answer questions about them? This model, llava-medical-8B-clip-vit-stage2, is a specialized vision-language model fine-tuned for healthcare. It's like giving a chatbot eyes for X-rays and MRIs! #AI# #MedTech#
Show more
Ever wanted a compact AI that talks like a pro? Meet Ornith-1.5-9B-GGUF, a text-generation powerhouse that fits right into your workflow. With 54k downloads and 101 likes, it's already winning hearts. Curious? Let's dive in! #AI# #LLM#
Show more
Ever wanted an AI that doesn't hold back? Meet Qwen3.8-27B-OBLITERATED, an abliterated model that's all about uncensored text generation. It's a red-team's dream, pushing boundaries while keeping the Qwen3.5 smarts. 4.4k downloads and counting, this is the edge you've been looking for. #AI# #LLM#
Show more
NVIDIA just dropped PersonaPlex-7B 🤯
A full-duplex voice model that listens and talks at the same time.
No pauses. No turn-taking. Real conversation.
100% open source. Free.
Voice AI just leveled up.
Show more
