🐱

Phate Phate334

🐱

Phate334 / slim-python.omp.json

Created January 31, 2023 08:03

oh-my-posh python style

	{
	"$schema": "https://raw.githubusercontent.com/JanDeDobbeleer/oh-my-posh/main/themes/schema.json",
	"blocks": [
	{
	"alignment": "left",
	"segments": [
	{
	"background": "#2f2f2f",
	"foreground": "#26C6DA",
	"leading_diamond": "\ue0c7\u2593",

Phate334 / Translated-Whisper.md

Last active January 22, 2025 17:01

Faster Whisper + Google translate

Install

!pip install faster-whisper
!pip install googletrans==3.1.0a0
!pip install pysubs2

Phate334 / llama.py

Created August 15, 2023 01:55

patch from https://github.com/abetlen/llama-cpp-python/pull/309 for v0.1.77

	import os
	import sys
	import uuid
	import time
	import math
	import multiprocessing
	from abc import ABC, abstractmethod
	from typing import (
	List,
	Optional,

Phate334 / backup-docker-compose.sh

Last active April 23, 2025 01:47

Simple script to backup volumes in docker compose env.

	#!/bin/bash
	set -euo pipefail

	# Function to get Docker API version
	get_docker_api_version() {
	docker version --format '{{.Server.APIVersion}}'
	}

	# Function to get volume paths from docker compose and backup
	backup_volumes() {

Phate334 / llamacpp-bench.sh

Last active May 13, 2025 03:10

A simple script to bench GGUF base and draft models

	#!/bin/bash
	set -e
	set -u

	if [ $# -lt 1 ]; then
	echo "Usage: $0 BASE_MODEL [DRAFT_MODEL] [--skip-base] [--skip-draft] [--skip-sd]"
	exit 1
	fi

	BASE_MODEL="$1"

Phate334 / compose.yaml

Last active May 16, 2025 09:22

Mistral 3.1 chat template for llama.cpp, including tools and support for multiple images. modify from vllm https://github.com/vllm-project/vllm/blob/main/examples/tool_chat_template_mistral.jinja

	services:
	llm:
	image: ghcr.io/ggml-org/llama.cpp:server-cuda-b5391
	ports:
	- "7071:8080"
	volumes:
	- ./models:/models
	env_file:
	- llm.env
	deploy: