DEDUPED Hot AI repos, turned into use cases Get 5 a day

How to build a high-fidelity pet with the Qwen identity backend with Pocket Men with You

Prepare the heavier local Qwen editor and create a max-quality pet that keeps the subject's look.

  • Content
  • Advanced
  • 2 hours
  • From the README
  • Any agent

Who it is forA user who cares more about identity fidelity than setup size or latency, and has a GPU that can run the heavier model.

The problemThe default backend may not hold a subject's likeness closely enough for a companion the user wants to look like them.

The outcomeA max-quality companion created with the Qwen-Image-Edit-2511 backend, with state-specific key poses for every state.

What you need first

  • Two or more reference photos of the subject (the README says raw personal references stay local by default)
  • A Windows machine for the .bat installers, which the README describes
  • Python, the repo language, for the pocketmen CLI
  • An NVIDIA GPU with enough VRAM for the neural engine; otherwise PocketMen falls back to the deterministic v0.2 engine
  • Codex, if you want to use the $pocketmen-with-you skill

Steps

  1. Run PREPARE_IDENTITY_MAX.bat to download and cache the Qwen-Image-Edit-2511 weights.
  2. Run pocketmen doctor to confirm the hardware can run the heavier backend.
  3. Run pocketmen create --reference photo-1.jpg --reference photo-2.jpg --engine neural --backend qwen-image-edit-2511 --quality max, adding the name, style, and output flags you need. (inferred, check the README)
  4. Check the output folder for pet.json and spritesheet.webp before installing.

Prompt for your agent

In the repo six-nut/PocketMen-with-you, run `pocketmen create` with --engine neural --backend qwen-image-edit-2511 --quality max using the references at <photo-1> and <photo-2>. Expected output: a pet.json and spritesheet.webp built from the Qwen backend. Verify that the run did not use the OpenAI API and that the atlas is 8 by 9 cells.

Caveat. The README says this backend is heavier than the default, so the model download and generation time may be long on consumer hardware.

Share: X LinkedIn Hacker News Email Copy link

Disclaimer

Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page