Skip to content
HearmemanAI

Image Generation

What is included. What to know before you deploy.

All templates
Image Generation. Generate and edit images. Choose the models you need. HearmemanAI.

Image Generation

Generate and edit images, explore camera angles and upscale your results. Choose the models you need before deploying.

Available models

  • MOSS-Image / Kromoss
  • Qwen Image 2.1
  • Qwen Image 2512
  • Qwen Image Edit 2511
  • Z-Image Turbo
  • Krea 2
  • Boogu
  • FLUX.3 Klein
  • ComfyUI and JupyterLab with selectable image model sets
  • Generation, editing, camera-angle and upscaling workflows
  • MOSS-Image SFT and an 8-step Kromoss option in one workflow
  • FLUX.3 Klein generation and editing with approved Hugging Face access

20 bundled workflows across 8 optional model sets

Runs the InfluencerMaxxing kit workflows out of the box. InfluencerMaxxing 101 · $69.99 Learn more

See the workflow breakdown
MOSS-Image / Kromoss 1 workflow
Text-to-image generation. Leave KROMOSS off for MOSS SFT, or turn it on for Kromoss with the 8-step LoRA.
Qwen Image 2.1 2 workflows
Text-to-image generation and image editing.
Qwen Image 2512 4 workflows
Generation variants, including a PiD upscaling graph.
Qwen Image Edit 2511 3 workflows
Image editing and multiple camera angles.
Z-Image Turbo 3 workflows
Generation, tiled upscaling and PiD upscaling.
Krea 2 2 workflows
RAW plus Turbo LoRA generation, with a PiD variant.
Boogu 3 workflows
Separate Base, Turbo and Edit graphs.
FLUX.3 Klein 2 workflows
Distilled INT8 text-to-image and image editing. Requires a Hugging Face token and approved model access.
Before you deploy

Before deploying, click Edit Template and open Environment Variables. Enable at least one model set by changing its download variable from false to true. For MOSS-Image and Kromoss, set DOWNLOAD_MOSS=true; both use public weights. For FLUX.3 Klein, accept the model license and obtain access on Hugging Face, then set DOWNLOAD_KLEIN3=true and HF_TOKEN. MOSS requires the current v18 image or a newer compatible image; restarting an older pod does not upgrade its ComfyUI installation.

Model selection

MOSS-Image / KromossDOWNLOAD_MOSSDefault: false
Enable if wanted false → trueSet true to download MOSS SFT INT8, Kromoss INT8, the Qwen3.5-9B W4A8 text encoder, Flux 2 small VAE and 8-step LoRA, and install the MOSS Image workflow. The KROMOSS toggle selects the sampling path. Requires ComfyUI 0.42.1 or newer; no Hugging Face access approval is needed.
Qwen Image 2.1download_qwen_21Default: false
Enable if wanted false → trueText-to-image generation and image editing. Change to true to download its models and install both workflows.
Qwen Image 2512DOWNLOAD_QWEN_IMAGEDefault: false
Enable if wanted false → trueImage generation.
Qwen Image Edit 2511DOWNLOAD_QWEN_IMAGE_EDITDefault: false
Enable if wanted false → trueImage editing and camera angles.
Z-Image TurboDOWNLOAD_Z_IMAGEDefault: false
Enable if wanted false → trueImage generation and upscaling.
Boogudownload_booguDefault: false
Enable if wanted false → trueBase, Edit and Turbo models.
Krea 2download_krea2Default: false
Enable if wanted false → trueRAW, Turbo and identity-edit LoRA downloads with two generation workflows.
FLUX.3 KleinDOWNLOAD_KLEIN3Default: false
Enable if wanted false → trueSet true to download the Klein 3 models and install its two bundled workflows. Requires HF_TOKEN and model access. A failed token or access check logs an error and skips only Klein 3; template setup continues.

OptionalModel precision

QWEN_IMAGE_PRECISIONDefault: bf16
Use bf16 or fp8 for Qwen Image 2512 generation. This does not change Qwen Image Edit or Qwen Image 2.1. Copied workflows use the selected files.
BOOGU_PRECISIONDefault: bf16
Use bf16 or fp8 for the Boogu models, independently of Qwen. Copied workflows use the selected files.

OptionalCivitAI downloads

civitai_tokenDefault: empty
Your CivitAI API token. Add it if you want to download your own LoRAs or checkpoints.
CIVITAI_LORASDefault: empty
Comma-separated LoRA model-version IDs to download into models/loras. Replace any replace_with_ids placeholder with your IDs.
CIVITAI_CHECKPOINTSDefault: empty
Comma-separated checkpoint model-version IDs to download into models/checkpoints. Leave empty if you do not need any.
How to download CivitAI LoRAs

Model access

For Klein 3, accept the model license and obtain access on Hugging Face, then create a read token from that same account. Set HF_TOKEN and DOWNLOAD_KLEIN3=true before deploying. Keep your token private.

HF_TOKENDefault: empty
Required for Klein 3; optional for other model families. Use a Hugging Face read token from the account that accepted the model license and obtained access. Fine-grained tokens must allow reading black-forest-labs/flux-3-image-klein-comfy. A token alone does not grant model access.
GITHUB_PATDefault: empty
Access token for the extra node packs used by purchased workflows. Not needed for the free workflows included in this template.
Deploy on Runpod

HearmemanAI may earn a commission when you use this Runpod link. GPU and storage charges apply.