New Models

ZwZ2-9B-GGUF Vision-Language Model: 8GB+ VRAM

GGUF format version of ZwZ2-9B, an open-weight vision-language model focused on spatial perception and evidence conne ...

Image, Video and Audio

Pika-H3-Flex Audio-Visual Video Generation Model: File List

Pika-Labs has released Pika H3 Flex, a model adapter that generates videos with synchronized stereo audio from text, ...

New Models

Ling-3.0-flash-Sante-fp8 Text Generation Model: ~144GB Memory

inclusionAI has released Ling-3.0-flash-Sante-fp8, an FP8 quantized medical LLM offering 256K context and strong heal ...

New Models

Mistral-Large-4-1T-A52B-NVFP4-GGUF Vision-Language Model: 8GB+ VRAM

Explore the GGUF release of Mistral AI’s Mistral Large 4 NVFP4 quantized model, featuring a 1M context window a ...

New Models

Mistral AI Releases Mistral Large 4 Open-Weight Model

Mistral AI has released model weights for Mistral Large 4 (Le Chonk), a 1-trillion-parameter native multimodal open-w ...

New Models

Mistral Large 4 Released: Multimodal MoE Model

Learn about Mistral Large 4, a multimodal multilingual MoE model by Mistral AI, its capabilities, hardware requiremen ...

Engines and Tools

Magnitude CLI 0.2.14 Released: Faster CPU Decoding and More

Version 0.2.14 of @magnitudedev/cli brings faster GGUF decoding on CPUs, improved download resilience, and bug fixes ...

Engines and Tools

Open WebUI v0.12.5 Released with Security Patches and Multi-Page Notes

Open WebUI v0.12.5 is now available, featuring important security fixes, multi-page notes, faster knowledge base sear ...

Image, Video and Audio

LTX-2-Marking Video and Audio Watermarking Model: File List

Learn about Lightricks/LTX-2-Marking, model weights and configuration files for invisible watermarking in the LTX-2 v ...

Image, Video and Audio

granite-speech-5.0-2b-ONNX Speech Recognition Model: 8GB+ VRAM

IBM has released ONNX weights for Granite Speech 5.0 2B, enabling in-browser ASR and AST inference via Transformers.j ...