mirror of
https://github.com/langgenius/dify-plugin-daemon.git
synced 2026-07-21 17:25:23 -04:00
Nano Banana pro cannot generate 4k images, Request to Plugin Daemon Service failed-500 #218
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @Valentino-L on GitHub (Dec 10, 2025).
Self Checks
Dify version
1.10.0
Cloud or Self Hosted
Self Hosted (Docker)
Steps to reproduce
Choose 4k option here
✔️ Expected Behavior
Successfully generate 4k images
❌ Actual Behavior
Always times out, dify is unable to output any images. Gives "Request to Plugin Daemon Service failed-500" error
@dosubot[bot] commented on GitHub (Dec 10, 2025):
Hi @Valentino-L! I'm Dosu and I’m helping the dify team.
4K image generation with Nano Banana Pro is not officially supported in Dify v1.10.0—only High and Low resolutions are stable. Selecting 4K currently has no effect and can cause issues, as confirmed by maintainers; 4K support is under development and not yet available in the release you’re using (source).
The “Request to Plugin Daemon Service failed-500” error with a ~300s timeout happens because the Plugin Daemon Service times out on long-running tasks like large image generation. This is controlled by the
PLUGIN_DAEMON_TIMEOUT(default 300s) and related environment variables (source). There are no explicit per-model or per-resolution limits—timeouts and resource exhaustion are the main causes (source).If you want to experiment with longer tasks, increase these environment variables in your
.envordocker-compose.yamland restart your Docker stack:PLUGIN_DAEMON_TIMEOUT=1200PLUGIN_MAX_EXECUTION_TIMEOUT=1200WORKFLOW_MAX_EXECUTION_TIME=1800HTTP_REQUEST_MAX_READ_TIMEOUT=1200GUNICORN_TIMEOUT=600NGINX_PROXY_READ_TIMEOUT=3600sNGINX_PROXY_SEND_TIMEOUT=3600sMake sure your Docker host has enough CPU, RAM, and (if needed) GPU resources—Dify does not set explicit resource limits, so you must manage these externally (source). Also, check Docker networking, service health, and Redis connectivity to rule out misconfiguration (source).
Until official 4K support lands, use High resolution for Nano Banana Pro. If you need 4K output immediately, consider alternative plugins (like Regolo.ai) or models with better streaming support (source). If you want help interpreting release notes or config changes when 4K support is released, let me know!
To reply, just mention @dosu.
How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other
@crazywoola commented on GitHub (Dec 10, 2025):
Issues related to the plugins should not be posted in this repo.
@Valentino-L commented on GitHub (Dec 10, 2025):
This is not a plugin, it's a problem with the official dify LLM node and Gemini is the model provider
@crazywoola commented on GitHub (Dec 10, 2025):
It seems to be a problem with this repo, please attached the logs as well.
@dosubot[bot] commented on GitHub (Dec 10, 2025):
The 500 error and timeout when generating 4K images with Nano Banana Pro are most likely caused by the plugin daemon's execution timeout or resource exhaustion in your Docker environment. For resource-intensive tasks like 4K image generation, you should increase the
PLUGIN_MAX_EXECUTION_TIMEOUT(default is 600 seconds) andGUNICORN_TIMEOUT(default is 200 seconds) environment variables in your Docker Compose or deployment configuration. This gives the plugin more time to complete before being killed by the daemon. Also, review and raise your container's CPU, memory, and (if used) GPU limits to ensure the plugin isn't being killed due to resource constraints. Timeout enforcement happens at the infrastructure level, so these changes are critical for long-running jobs like 4K renders. Examining the plugin daemon logs around the time of the error can provide more details on the failure cause. See more details in the relevant discussion and the .env.example configuration.To reply, just mention @dosu.
How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other