- Add verify-env.sh script to test .env file loading
- Verify GOOGLE_API_KEY is accessible in container
- Check docker-compose config and container environment
- Ensure Python can access environment variables via os.environ
- Provides reproducible verification process for development
- Add comprehensive RunPod deployment guide
- Document environment variable setup in RunPod template UI
- Explain differences between local and RunPod configurations
- Include security notes and verification steps
- Fix venv creation check to look for bin/activate instead of directory
- Add INSTALL_NANO_BANANA env var to entrypoint for automatic installation
- Add COMFYUI_ARGS support to pass arguments to ComfyUI (e.g., --cpu)
- Configure CPU mode for testing without GPU
- Update comfy-node-install to disable git terminal prompts
Note: Nano Banana repository URL needs verification (current URL returns 404)
- Updated PLAN.md with RunPod specific configuration instructions
- Confirmed RunPod uses template env vars (os.environ)
- No code changes needed for RunPod compatibility (Nano Banana supports env vars natively)
- Add Dockerfile with CUDA 12.6 and Python 3.12 base
- Add docker-compose.yml with GPU support and volume mounts
- Add docker-entrypoint.sh script with venv management and OTEL support
- Add comfy-node-install.sh helper for custom node installation
- Update .gitignore to exclude Docker-related artifacts
- Entrypoint ensures frontend package matches requirements.txt version
- Optional OTEL instrumentation when OTEL_EXPORTER_OTLP_ENDPOINT is set
- Volume-mounted repo allows git checkout of release tags without rebuilds
From now on ComfyUI will do version numbers a bit differently, every stable
off the master branch will increment the minor version. Anytime a fix needs
to be backported onto a stable version the patch version will be
incremented.
Example: We release v0.6.0 off the master branch then a day later a bug is
discovered and we decide to backport the fix onto the v0.6.0 stable, this
will be done in a separate branch in the main repository and this new
stable will be tagged v0.6.1
* make setattr safe for non existent attributes
Handle the case where the attribute doesnt exist by returning a static
sentinel (distinct from None). If the sentinel is passed in as the set
value, del the attr.
* Account for dequantization and type-casts in offload costs
When measuring the cost of offload, identify weights that need a type
change or dequantization and add the size of the conversion result
to the offload cost.
This is mutually exclusive with lowvram patches which already has
a large conservative estimate and wont overlap the dequant cost so\
dont double count.
* Set the compute type on CLIP MPs
So that the loader can know the size of weights for dequant accounting.
In the lowvram case, this now does its math in the model dtype in the
post de-quantization domain. Account for that. The patching was also
put back on the compute stream getting it off-peak so relax the
MATH_FACTOR to only x2 so get out of the worst-case assumption of
everything peaking at once.
