5.3.0

5.3.0 (2025-07-22)

Features

• add support for 5090/B200 (#150) (7492513)

5.2.0

5.2.0 (2025-07-02)

Features

• update comfyui to 0.3.43 (#145) (32c2395)

5.1.1

5.1.1 (2025-07-01)

Bug Fixes

• update runpod to 1.7.12 (#144) (c689eda)

5.1.0

5.1.0 (2025-05-28)

Bug Fixes

• add missing libs needed for "opencv" to work (33cd5d3)
• use proper example that actually can be copied & pasted (506fac1)

Features

• added COMFY_LOG_LEVEL to control the logs from ComfyUI (183de1c)
• added script "comfy-manager-set-mode" to set the "network mode" (4d5425f)
• added script "comfy-node-install" as a wrapper to "comfy-cli" to proivde a proper exit code when something is not corret (6605506)
• improved logging & provide actual reason why the workflow validation might fail (7582999)
• updated to ubuntu 24.04 & python 3.12 to support ffmpeg 6 (eb8dd26)

5.0.4

• trigger a new release on the hub

5.0.3

• trigger a new release on the hub

5.0.2

• ci: trigger tests

5.0.1

• test: simplified workflow & use it for both tests

5.0.0

Breaking Change

• Versions < 5.0.0 returned the primary image data (S3 URL or base64 string) directly within an output.message field
• Starting with v5.0.0, the output format has changed significantly:
  • Image data is now returned within an output.images field
  • This field is a list of dictionaries. Each dictionary contains:
    • filename: The original filename from ComfyUI
    • type: Either "s3_url" or "base64"
    • data: The corresponding S3 URL or the base64 encoded image string
• Clients need to be updated to handle this new structure as it's not compatible with previous versions.

Features

• Websocket Communication: Replaced HTTP polling with websockets for job status monitoring and completion.

  • fixes #108
  • fixes #103

• Multiple Output Images: Processes and returns all non-temp images from the workflow history

  • fixes #95
  • fixes #45

• Simplified Customization Workflow: Users no longer need to fork the repository & use the "snapshot restoration" workflow to add custom nodes or models. They can create their own Dockerfile that builds FROM an official base image, using comfy-cli for setup to install custom-nodes and download models

  • fixes #116
  • fixes #107
  • fixes #109
  • fixes #97

• ComfyUI update: 0.3.30

• Direct Image Retrieval: Fetches image data via ComfyUI /view API instead of filesystem access

• Filename in Output: Includes original filename in the output data for each image

• Websocket Reconnection: Added automatic reconnection attempts for dropped connections during monitoring

• Unhandled Output Warnings: Logs warnings for unrecognized node output types

• Removed Redundant Output: Removed the refresh_worker key from handler.py return values, as the behavior is solely controlled by the environment variable

Documentation

• Major Restructure: Refactored the main README.md into a concise overview and quickstart guide, moving detailed sections (deployment, configuration, customization, development, ci/cd, acknowledgments) into separate, focused files within the docs/ directory for improved readability and maintainability
• Improved Customization Guide: Rewrote docs/customization.md to feature the new, simpler method using a custom Dockerfile with comfy-cli commands (comfy node install, comfy model download). Removed the old method requiring repository forks and snapshots

Bug Fixes

• Input Image Base64 Decoding: Correctly handles optional data:[...];base64,... prefixes in input image strings.

4.1.0

4.1.0 (2025-05-02)

Bug Fixes

• moved code back into stage 1 (d9ed145)

Features

• move stuff around (2b2bc12)
• removed polling; added websocket; allow multiple output images; (79a560f)