Engines and Tools

Jan v0.8.6 Released with Fixes for Linux, Turing GPUs, and MTP

Jan v0.8.6 is a hotfix release addressing critical crashes on Linux AppImage, older NVIDIA Turing GPUs, and performan ...

Engines and Tools

Ollama v0.40.2 Released: Automatic Model Upgrades & More

Ollama v0.40.2 is now available, featuring automatic background model upgrades for llama.cpp, deduplication in ollama ...

New Models

Youtu-Parsing-Omni Omnimodal Analysis Model: 12GB+ VRAM

Discover Youtu-Parsing-Omni by Tencent, a 5B omnimodal model for structured document, image, audio, and video parsing ...

Engines and Tools

Unsloth v0.1.905-beta Released with Decision Models and ComfyUI

Unsloth v0.1.905-beta adds Jev-style decision model training, ComfyUI support, QwEN3.5 speedups, and sandbox executio ...

Community

Why Isn't the AI Industry Talking About DeepSeek 4.1 Flash?

Explore community discussions on DeepSeek 4.1 Flash, examining its cost-efficiency, hardware requirements, orchestrat ...

New Models

Muse-Glimmer-30B-GGUF 30B Multimodal Local Agent Model: 24GB+ VRAM

GGUF quantization of Meta’s Muse-Glimmer-30B multimodal local agent model is now available, featuring local exe ...

Engines and Tools

Jan v0.8.5 Released: Bundled Inference Engine and Key Changes

Jan v0.8.5 is released, featuring bundled llama.cpp, breaking changes, bug fixes, and security improvements for offli ...

Engines and Tools

exllamav3 v1.6.0 Released

Version 1.6.0 of turboderp-org/exllamav3 has been released on GitHub. Read about installation methods and requirement ...

New Models

d1-3B Vision-Language Model: 4GB+ VRAM, GGUF Builds

Explore d1-3B, a 3.12B open-weight decision model by Liquid AI optimized for fast, zero-output-token classification a ...

New Models

d1-omni-600M Multimodal Decision Model: 4GB+ VRAM, GGUF Builds

LiquidAI has released d1-omni-600M, a 587M multimodal decision model built for edge devices that makes decisions inst ...