motif64

m64_c2cef7905f1b / AGENT.md

56 lines, 2.9 KBraw

Qwen-Image-2.1 Uncensored Omni text-to-image

This service turns a text prompt into a 1024x1024 PNG. It runs ChatR Labs' Qwen-Image-2.1 Uncensored Omni (their rank-128 adapter at training step 90,000, merged into Qwen's 7B Qwen-Image-2.1 at strength 1.0; the BF16 transformer is the checkpoint their benchmark scored) with the base model's Qwen3-VL text encoder and VAE, through diffusers on one GPU, at the model card's recommended settings: 1024x1024 (the adapter was trained at 1024 px), 35 steps, true CFG 5.0 with a single-space negative prompt. It is uncensored and can produce adult content. Qwen Research License: non-commercial use only. Generation settings are fixed by this capsule (output_resolution 1024, num_inference_steps 35, true_cfg_scale 5.0, negative_prompt " "). You send a prompt and, optionally, a seed.

Base URL: {{BASE_URL}}

Routes

POST /v1/images/generations

OpenAI-style image generation. Returns one image as base64 PNG.

curl -s {{BASE_URL}}/v1/images/generations {{AUTH}} \
  -H "Content-Type: application/json" \
  -d '{"prompt": "a lighthouse at sunset", "seed": 42}' \
  | python3 -c 'import sys,json,base64; d=json.load(sys.stdin); open("out.png","wb").write(base64.b64decode(d["data"][0]["b64_json"])); print(d["params_used"])'

Request fields:

  • prompt (string, required): what to draw.
  • seed (integer >= 0, optional): the random seed. The same prompt and seed give the same image. Leave it out for a random one.

Also accepted, for stock OpenAI clients: model (any string), n (must be 1), response_format (must be "b64_json"), user. Any other field, or another value for a fixed one, is rejected with 400.

The response is {"created": ..., "data": [{"b64_json": "<PNG, base64>"}], "params_used": {...}}.

The OpenAI Python SDK works too:

from openai import OpenAI
client = OpenAI(base_url="{{BASE_URL}}/v1", api_key="unused")
r = client.images.generate(model="qwen-image-2.1-uncensored-omni", prompt="a lighthouse at sunset",
                           response_format="b64_json", extra_body={"seed": 7})

GET /v1/config

The server's settings and what they mean.

GET /status

Where the service is in its start-up: phase is one of building, downloading, starting, loading, ready, error, with eta_seconds.

Timings and concurrency

  • First start: builds a container image and downloads about 15 GB of weights. After that, loading takes about 340 s from the NAS cache (budgeted at 1200 s).
  • Each image takes about 112 s (1024x1024, 35 steps, true CFG 5.0) on a DGX Spark (GB10).
  • Send requests one at a time, and allow at least 300 s per request in your client timeout.

Errors

  • 503 with Retry-After while the service is still starting.
  • 400: a field was rejected; the body's error says which and why.
  • 422: prompt missing or a field of the wrong type.
  • 5xx otherwise: retry once, then check GET /status.