Post

AI Gen Models

Posted Updated
By Aaryan Poudel 1 min read

Music Generation

1. Ace-step

  • Open Source. Github
  • Tried on M1 Air 8GB. During generation, ran into out of memory error as the swap usage exceeded 9GB

Setup:

1
2
3
4
5
6
7
8
9
10
11
12

git clone https://github.com/ace-step/ACE-Step.git
cd ACE-Step
python -m venv .venv
source .venv/bin/activate

pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu
# Or, for Nvidia 
# pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126

pip install -e .

acestep --port 7865

2. Suno.ai

  • Paid proprietary music gen
  • Created great songs but I think it’s free tier is for non commercial use

Image Editing

1. Qwen Image Edit

  • Open Source. Huggingface
  • Semantic Image Editing / Image editing using prompts
  • Can also try via Qwen Chat

2. Flux Kontext Dev

  • Open Source (verify if OSS or Open Weight) Huggingface
  • Semantic Image Editing

3. Nano Banana

  • Gemini 2.5 Flash Image Preview model as the image editor
  • Can be used via AIStudio

Video Generation

  1. Wan2.2

Tools

1. Comfy UI

2. LM Studio

  • Download
  • Models Currently Using in LMStudio
  1. liquid-lmf2
  2. qwen2.5-coder
  3. gemma-3
Blogs
blog
This post is licensed under CC BY 4.0 by the author.