Feature request "Start a new chat" with a summary of the context of the current one

DDR6 RAM and a reasonable GPU should be able to run 70b models with good speed

MiniMax-Text-01 - A powerful new MoE language model with 456B total parameters (45.9 billion activated)

22 startup ideas to start in 2025 (ai agents, saas, etc)

Kokoro #1 on TTS leaderboard

6x AMD Instinct Mi60 AI Server vs Llama 405B + vLLM + Open-WebUI - Impressive!

Ag, Claude feels a bit weird today... Are we about to get a new model?

LocalGLaDOS - running on a real LLM-rig

New Model from https://novasky-ai.github.io/ Sky-T1-32B-Preview, open-source reasoning model that matches o1-preview on popular reasoning and coding benchmarks — trained under $450!

Does anyone know how to replicate this setup for coding ?

New Moondream 2B vision language model release

Windows compatibility layer Wine 10.0 third Release Candidate out now

RTX 4090 48GB - $4700 on eBay. Is it legit?

What is the best LLM for writing and coding that can be run on 4090?

Do you think you can find the password ? I made a small LLM challenge

How do you keep up with the SOTA of everything? Where's the best leaderboards?

Oh shit

DeepSeek V3 GGUF 2-bit surprisingly works! + BF16, other quants

Comparing AGI safety standards to Chernobyl: "The entire AI industry is uses the logic of, "Well, we built a heap of uranium bricks X high, and that didn't melt down -- the AI did not build a smarter AI and destroy the world -- so clearly it is safe to try stacking X*10 uranium bricks next time."

OpenAI o1 playing chess against 4o

Now THIS is interesting

"Quick update: We’re making steady progress, with send-outs, delayed deposits, and trade issues being resolved..."

I'm sorry WHAT? AMD Ryzen AI Max+ 395 2.2x faster than 4090

Announcement made at AMD at CES 2025 - New Ryzen CPU (AMD Ryzen AI max+ 395) for laptop runs a 70B(q4) 2 times faster than a 4090 discrete desktop GPU

RTX 5090 rumored to have 1.8 TB/s memory bandwidth