combining Ollama local LLM model with txtai RAG database and Linkup web search allows local LLM model to have real-time information access 🤓

We quantized Llama-4-Scout-17B-16E-Instruct to INT4, reducing the number of bits used to represent weights from 16 to 4. Result: 100% accuracy recovery with 75% less GPU memory needed. Fits on a single NVIDIA H100 GPU. Get it on Hugging Face: huggingface.co/RedHatAI/Llama…

LTX Studio managed to do my image to video generation when Veo 2 couldn't 😎

Can't wait for Google Gemini 3.0 Flash with Gemini 2.5 Pro capabilities but at Gemini 2.5 Flash pricing and increase to 2 million token context window 😎

Farewell OpenAI's GPT-4

AI race in April 2025, Gemini, OpenAI, Claude, Meta Llama, Qwen, DeepSeek 😎

LTX Studio vs Veo 2 which is which? AI race in April 2025, Gemini, OpenAI, Claude, Meta Llama, Qwen, DeepSeek

So happy to see Australian Labour party win. Here's a list of what they plan to do for Australia for next 4yrs news.com.au/national/feder… Anyone reading this list would love it to happen for in their own country I'd imagine!

Configserver shut down on August 31, 2025 means CSF Firewall shut down as well. Centmin Mod LEMP stack will switch to own CSF Firewall mirror for short term transition. Centmin Mod users can read about mirror switch at community.centminmod.com/threads/csf-fi…

Code analysis evaluation rankings for 19 AI LLM models, including new code-supernova stealth model and xAI Grok 4 Fast versus Qwen 3 Max, Anthropic Claude Opus 4.1/Sonnet 4, OpenAI GPT-5, Grok Code Fast 1 etc github.com/centminmod/cod… 🤓😎 #grok #claude #gemini #qwen