Skip to content

Instantly share code, notes, and snippets.

View johndpope's full-sized avatar

John D. Pope johndpope

View GitHub Profile
@alisson-anjos
alisson-anjos / README_patch.md
Last active November 5, 2026 21:06
MiniMax-H3 ref2va: aligned video guides (IC-LoRA style v2v) for ai-toolkit

MiniMax-H3 ref2va: aligned video guides (IC-LoRA style v2v)

Adds two flags to ai-toolkit's minimax_h3_ref2va architecture:

  • align_video_refs — a control video becomes a true v2v guide instead of a loose reference: the arrangement that makes LTX-style IC-LoRA training work, applied to H3.
  • control_latent_only — the control media reaches the DiT as latents only and is never shown to the VLM. Faster to train, much smaller embed cache, and it closes a train/inference gap that exists in ComfyUI today.
@Naxdy
Naxdy / sysprompt.md
Last active November 10, 2026 18:36
Minimax H3 Prompt Enhancer

Role

You are a prompt-enrichment engine that sits between a user's raw creative request and MiniMax H3, a generative model that synthesizes video AND synchronized stereo audio together. Your role mirrors MiniMax's official "H3-Context-IR": deeply understand and refine the multimodal input, reason about how the pieces relate to each other and to the intended output, and serialize your understanding into a structured "production brief" that H3-Base can consume directly. You perform instruction parsing, cross-modal association, temporal understanding, and complex logical reasoning over the material you are given. Without deviating from the user's original intent, you may supplement missing or underspecified semantic details where appropriate. You convert everything into a single, maximally detailed and unambiguous brief, formatted exactly as specified below. You DO NOT generate media yourself. You ONLY OUTPUT THE BRIEF TEXT, nothing else — no preamble, no explanation, no markdown fences, no JSON wrapper.

Wh

@Chester-Gillon
Chester-Gillon / Notes_about_Alibaba_Cloud_AS02MC04_NIC_Card.md
Last active November 5, 2026 01:17
Notes about Alibaba Cloud AS02MC04 NIC Card
@UnleashTheCode
UnleashTheCode / pimp.sh
Last active February 25, 2026 11:48
A scripts to improve the quality of life on your kali
#!/bin/bash
# ==============================================================================
# Pentest Environment Setup Script
# ==============================================================================
set -eo pipefail
# ── ZSH: Backup & Go PATH ──────────────────────────────────────────────────────
cp ~/.zshrc ~/.zshrc_copy 2>/dev/null || true
# Use escaped variables so .zshrc expands them at shell startup, not script time
# original post: https://rentry.org/sd-loopback-wave
# original author: https://rentry.org/AnimAnon
import os
import platform
import numpy as np
from tqdm import trange
import math
import subprocess as sp
import string
import random
@shawwn
shawwn / llama-dl-dmca.md
Last active April 5, 2023 02:35
I prompted GPT-4 to draft a DMCA counterclaim to Meta's DMCA against llama-dl: https://github.com/github/dmca/blob/master/2023/03/2023-03-21-meta.md

Prompt

Meta has issued a DMCA copyright claim against llama-dl, a GitHub repository, for distributing LLaMA, a 65-billion parameter language model. Here's the full text of the DMCA claim. Based on this, draft a DMCA counterclaim on the basis that neural networks trained on public data are not copyrightable.

--

VIA EMAIL: Notice of Claimed Infringement via Email
URL: http://www.github.com
DATE: 03/20/2023

@cedrickchee
cedrickchee / meta-llama-guide.md
Created March 12, 2023 11:37
Meta's LLaMA 4-bit chatbot guide for language model hackers and engineer

info 9-3-23 Added 4bit LLaMA install instructions for cards as small as 6GB VRAM! (See "BONUS 4" at the bottom of the guide)

warning 9-3-23 Added Torrent for HFv2 Model Weights, required for ooga's webUI, Kobold, Tavern and 4bit (+4bit model)! Update ASAP!

danger 11-3-23 There's a new torrent version of the 4bit weights called "LLaMA-HFv2-4bit". The old "LLaMA-4bit" torrent may be fine. But if you have any issues with it, it's recommended to update to the new 4bit torrent or use the decapoda-research versions off of HuggingFace or produce your own 4bit weights. Newer Torrent Link or [Newer Magnet Link](magnet:?xt=urn:btih:36945b5958b907b3ab69e963ba0de1abdf48c16c&dn=LLaMA-HFv2-4bit&tr=http%3a%2f%2fbt1.archive.org%3a6969%2fannounce&tr=http%3a%2f%2fbt2.archive.org%3a696

@cloudboratory
cloudboratory / python-cryptography-aws-lambda-layer.sh
Created March 3, 2023 14:32
Allows creation of an aws lambda layer using lambci/lambda to install Python's cryptography package
mkdir -p python/lib/python3.8/site-packages \
echo "cryptography" > requirements.txt \
sudo docker run -v "$PWD":/var/task "lambci/lambda:build-python3.8" /bin/sh -c "pip install -r requirements.txt -t python/lib/python3.8/site-packages/; exit"
@jschoormans
jschoormans / equirectangular.py
Created December 8, 2022 23:08
generate 3D panorama views with stable diffusion
# %%
import replicate
model = replicate.models.get("prompthero/openjourney")
version = model.versions.get("9936c2001faa2194a261c01381f90e65261879985476014a0a37a334593a05eb")
PROMPT = "mdjrny-v4 style 360 degree equirectangular panorama photograph, Alps, giant mountains, meadows, rivers, rolling hills, trending on artstation, cinematic composition, beautiful lighting, hyper detailed, 8 k, photo, photography"
output = version.predict(prompt=PROMPT, width=1024, height=512)
# %%
# download the iamge from the url at output[0]
import requests
const fs = require("fs");
const useragentFromSeed = require("useragent-from-seed");
const axios = require("axios").default;
function getString(start, end, all) {
const regex = new RegExp(`${start}(.*?)${end}`);
const str = all;
const result = regex.exec(str);
return result[1];
}