adamteale

🚀 The 2026 M4 Max Local-First AI Guide (v2 — llama.cpp Edition)

Ditching Ollama: Direct llama.cpp for 2.4x Faster Inference

This document is the updated setup for a high-performance, local AI development environment on an M4 Max (36GB RAM). It replaces the Ollama backend with llama-server (upstream llama.cpp), delivering a 2.4x speedup on the fast-coder slot and 1.9x on the think-brain slot, with correct model templating and full control over every flag.

What changed from v1: Ollama is gone as the inference backend. llama-swap stays as the proxy but now talks directly to llama-server instead of spawning Ollama processes. Every flag you set now actually applies.

---

SolveSoul

[CONST_PARAM]

rtsp			= 1	;  RTSPЭ�飬0���ر�     1������

danilvalov

templates:
  global:
    transmission:
      host: xxxxxxxxxxxx
      port: xxxx
      username: xxxxxxx
      password: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

  tvshows:
    set:

stqism

#!/usr/bin/env sh

#Compile other/fun/sign.c from ProjectTox-Core and put it here named sign
#generate a keypair with ./sign g
#sh dns2key.sh v=tox1;etc {signing key}

  DNS=$1
  TOXVER=`echo -n $DNS | tr ';:"\\\/' '\n' | grep tox | tr '=' ' ' | awk '{print $NF}'`
  case $TOXVER in
    tox1)