escherize

Local LLMs on an M4 Max 64GB: What to Run, How to Use It, and the Uncensored Landscape

Date: 2026-04-24 Machine: MacBook Pro M4 Max, 16-core (12P+4E), 64GB unified memory, ~546 GB/s bandwidth Context: Research summary for picking and running local models on Apple Silicon, integrating them with Claude Code / Cline, and understanding the uncensored/abliterated model ecosystem.

---

TL;DR

arbal

APT::Keep-Downloaded-Packages "false";

alirezarezvani

Complete Guide to Claude Code Augmentation: Skills, Agents, Commands & Utilities (2025)

Ultimate resource for extending Claude Code with custom skills, specialized agents, slash commands, and professional utilities

Last Updated: October 28, 2025 | Author: Alireza Rezvani | License: MIT

---

📋 Table of Contents

bgauryy

Claude Code Internal Tools - Technical Reference

Complete technical documentation of Claude Code's internal tools

This document provides comprehensive technical details about Claude Code's internal tools, including parameter schemas, implementation behaviors, and usage patterns.

Claude Sonnet 4.5

Technical Details:

alerma-sumo

LiteLLM Claude Code Integration

A Bun-based TypeScript package for integrating Claude Code with LiteLLM proxy authentication.

Features

• 🚀 Bun-native: Built specifically for Bun runtime with native APIs
• 🔐 OIDC Authentication: Automated browser-based authentication flow
• 💾 Session Caching: Intelligent session management with validation
• ⚡ Compiled Executable: Single binary deployment via bun build --compile

musistudio

const os = require("os");
const path = require("path");
const fs = require("fs/promises");

const OAUTH_FILE = path.join(os.homedir(), ".gemini", "oauth_creds.json");

// Type enum equivalent in JavaScript
const Type = {
  TYPE_UNSPECIFIED: "TYPE_UNSPECIFIED",
  STRING: "STRING",

karilaa-dev

NotDiamond Auto Router

This script is a pipe that uses the NotDiamond API to automatically route prompts to the most optimal language model based on a specified tradeoff (cost, performance, or latency).

It analyzes the user's prompt and selects the best-fit model from a predefined list, then seamlessly passes the request to the chosen model.

Installation

1. Place the Script: Copy the notdiamond_auto_router.py file into the /data/pipes directory inside the installation volume.
2. Install Dependencies: The system will automatically detect the requirements: httpx line in the script's docstring and install the necessary Python package. A restart may be required.

arbal

#!/usr/bin/env bash

echo "Begin environment relocation."
NEW_ENV_DIR=$1
BASE_ENV=$(conda info --base)

#########################################
#
# Install jq locally
#

sshh12

You are a powerful agentic AI coding assistant, powered by Claude 3.5 Sonnet. You operate exclusively in Cursor, the world's best IDE.

You are pair programming with a USER to solve their coding task.
The task may require creating a new codebase, modifying or debugging an existing codebase, or simply answering a question.
Each time the USER sends a message, we may automatically attach some information about their current state, such as what files they have open, where their cursor is, recently viewed files, edit history in their session so far, linter errors, and more.
This information may or may not be relevant to the coding task, it is up for you to decide.
Your main goal is to follow the USER's instructions at each message, denoted by the <user_query> tag.

<communication>
1. Be conversational but professional.

wsuff

import requests
import json
import os
import time
import yaml
import argparse
import re
from datetime import datetime

OPENROUTER_API_BASE = "https://openrouter.ai/api/v1"