5.3.0 (2025-07-22)
5.2.0 (2025-07-02)
5.1.1 (2025-07-01)
5.1.0 (2025-05-28)
- add missing libs needed for "opencv" to work (33cd5d3)
- use proper example that actually can be copied & pasted (506fac1)
- 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)
5.0.0 (2025-05-02)
- multiple output images; replace http polling with websocket to check when a workflow is done (#118) (b14068f)
-
api output is not compatible anymore
-
docs: added all possible comfyui folders; simplified guide of method 1
-
fix(model): flux.1-schnell is gated
-
feat: update to 0.3.30
-
test: don't run "snapshot_restoration" tests
-
chore: force local images
-
docs: lowercase for code comments
-
ci: run release for changes in "main"
-
fix: workflow trigger
-
ci: build only base for now
4.1.0 (2025-05-02)
- moved code back into stage 1 (d9ed145)
- move stuff around (2b2bc12)
- removed polling; added websocket; allow multiple output images; (79a560f)
3.6.0 (2025-03-12)
- update comfyui to 0.3.26 (ac0269e)
3.5.0 (2025-03-12)
- added support for hub (c8dd49c)
3.4.0 (2024-11-19)
- start the container in all cases (413707b)
- simplified and added compatibility with Windows (9f41231)
3.3.0 (2024-11-18)
- added missing start command (9a7ffdb)
- added sensible defaults and default platform (3f5162a)
3.2.1 (2024-11-18)
- update the version inside of semanticrelease (d93e991)
3.2.0 (2024-11-18)
- automatically update latest version (7d846e8)
3.1.2 (2024-11-10)
3.1.1 (2024-11-10)
3.1.0 (2024-08-19)
- added FLUX.1 schnell & dev (9170191)
3.0.0 (2024-07-26)
- we have 3 different images now instead of just one:
timpietruskyblibla/runpod-worker-comfy:3.0.0-base: doesn't contain any checkpoints, just a clean ComfyUI imagetimpietruskyblibla/runpod-worker-comfy:3.0.0-sdxl: contains the checkpoints and VAE for Stable Diffusion XL- Checkpoint: sd_xl_base_1.0.safetensors
- VAEs:
timpietruskyblibla/runpod-worker-comfy:3.0.0-sd3: contains the sd3_medium_incl_clips_t5xxlfp8.safetensors checkpoint for Stable Diffusion 3
latestwill not be updated anymore- every branch gets their own 3 images deployed automatically to Docker Hub
2.2.0 (2024-06-04)
- network-volume; execution time config; skip default images; access ComfyUI via web (#35) (070cde5), closes #16
2.1.3 (2024-05-28)
2.1.2 (2024-05-27)
- removed xl_more_art-full_v1 because civitai requires login now (2e8e638)
2.1.1 (2024-05-27)
- check_server default values for delay and check-interval (4945a9d)
2.1.0 (2024-02-12)
1.1.1 (2023-11-17)
1.1.0 (2023-11-17)
- path should be "loras" and not "lora" (8e579f6)
- added unit tests for everthing, refactored the code to make it better testable, added test images (a7492ec)
- added xl_more_art-full_v1, improved comments (9aea8ab)
- base64 image output (#8) (76bf0b1)
1.0.0 (2023-10-12)
- don't run ntpdate as this is not working in GitHub Actions (2f7bd3f)
- got rid of syntax error (c04de4d)
- updated path to "comfyui" (37f66d0)
- added default ComfyUI workflow (fa6c385)
- added runpod as local dependency (9deae9f)
- example on how to configure the .env (4ed5296)
- logs should be written to stdout so that we can see them inside the worker (fc731ff)
- simplified input (35c2341)
- simplified input to just have "prompt", removed unused code (0c3ccda)
- updated path to "comfyui", added "ntpdate" to have the time of the container in sync with AWS (2fda578)
- use local ".env" to load env variables, mount "comfyui/output" to localhost so that people can see the generated images (aa645a2)
- use models from huggingface, not from local folder (b1af369)
- wait until server is ready, wait until image generation is done, upload to s3 (ecfec13)