Generate a new character image from an uploaded reference picture and an editable prompt. This ComfyUI graph uses MOSS-Image Base with the connected CharRef v2 LoRA and a MOSS image edit conditioning path.
Main features:
- Active model: moss_image_base_int8_convrot.safetensors with moss_image_charref_v2.safetensors at strength 1; qwen3.5_9b_bf16.safetensors and flux2-vae.safetensors are also loaded.
- The reference image passes through LoadImage and FluxKontextImageScale before MOSS image edit text conditioning.
- Output canvas 832 x 1216; Euler sampler at 50 steps, CFG 3, sgm_uniform scheduler, fixed seed 42. The graph saves one image.
Suggested workflow:
1. Install the named model, LoRA, text encoder, VAE, and required MOSS edit nodes; load the JSON.
2. Replace the example LoadImage input with your own character reference, then revise the positive prompt to match that character.
3. Generate and compare face, clothing, pose, and style against the reference. Character reference does not guarantee identical identity.
The supplied prompt includes an appearance description of a different braided character, so edit it before using a new reference. The paired cover is promotional artwork, not a measured output of this JSON.
RunningHub AI workflow: https://www.runninghub.ai/zh-cn/post/2114399090003243009?inviteCode=rh-v1111
RunningHub signup: https://www.runninghub.ai/?inviteCode=rh-v1111
Bilibili video: https://www.bilibili.com/video/BV1UYnU6qEGK/
Related model files: https://pan.quark.cn/s/9ec095b95838