LocalLLaMA !localllama@sh.itjust.works

Welcome to LocalLLaMA! Here we discuss running and developing machine learn

---

Read more

Members

3,369

Posts

353

Active Today

36

Created

2 yr. ago

Moderators

---

My AI Skeptic Friends Are All Nuts

Despite record gaming revenues, Nvidia reportedly plans to cut RTX 50 series production to allocate it to new AI hardware

Noob experience using local LLM as a D&D style DM.

Don't overlook llama.cpp's rpc-server feature.

LLMs and their efficiency, can they really replace humans?

DeepSeek just released updated r1 models with 'deeper and more complex reasoning patterns'. Includes a r1 distilled qwen3 8b model boasting "10% improved performance" over original

A gallery that showcases on-device ML/GenAI use cases and allows people to try and use models locally on your phone. APK

Using local model with basic RAG to help reference rules when playing table top game

How Hungry is AI? Benchmarking Energy, Water, and Carbon Footprint of LLM Inference

Has your local thinking model had an 'Aha!' moment similar to the one in Deepeek R1 papers?

mistralai/Devstral-Small-2505 · Hugging Face

I don't want them to, but Snapchat would have a great dataset to make an actually funny LLM.

Announcing Gemma 3n preview: powerful, efficient, mobile-first AI

Arch (btw), ROCm (AMD), docker (podman), llama.cpp (server) setup

Three Agent Protocols to Complete the Stack

Intel launches $299 Arc Pro B50 with 16GB of memory, 'Project Battlematrix' workstations with 24GB Arc Pro B60 GPUs

Anthropic's 'On the Biology of a LLM' got a massive update: Features fascinating deep dives into how models process information behind the scenes

StackOverflow activity down to 2008 numbers

8B Byte Latent Transformer model released by meta

Continuous Thought Machines