🧠Sogni Fast Worker Advanced Configuration

This page outlines additional settings you can tweak with your Sogni Fast Worker Install

Setting Your .env File Preferences

These first two options are available for the Stable Diffusion worker and automatically set during the worker-install process but you can also update them manually through the .env file. If changed manually you'll need to restart the worker to apply the changes either through the .bat / .sh files or by running docker compose down and then docker compose up from the command line.

AUTO_DOWNLOAD_TO_MIN_MODEL_COUNT=6

The Stable Diffusion worker comes bundled with 1 model by default (Sogni Artist SD1.5) and all 14 supported Controlnets. You may set the number of models you want your worker to download right away when it starts. Set this to 0 to disable automatic runtime model downloads.

While the count of models you are hosting is less than AUTO_DOWNLOAD_TO_MIN_MODEL_COUNT the worker will automatically download new models one at a time until this count of models is met. Models are downloaded based on real-time model popularity on the network with some models being 2gb and others being 6.7gb.

After this count is met, if your model folder size is still less than MAX_MODEL_FOLDER_SIZE_GB and AUTO_DOWNLOAD_TO_MIN_MODEL_COUNT is greater than 0 the worker will continue to automatically download new models but only as they are requested by the network. Sogni Supernet will only request new model downloads when the job queue for a model exceeds the amount of workers available to support it.

Models are cached on your local machine the first time they are downloaded so you won't have to download them again, unless you delete the models folder.

MAX_MODEL_FOLDER_SIZE_GB=300

Set the maximum size for your models folder in GB when using the Stable Diffusion worker. This includes Controlnets and other models. Any automated model downloads you configure with AUTO_DOWNLOAD_TO_MIN_MODEL_COUNT will stop before this limit is reached.

DATA_DOG_API_KEY=

Input the DATA_DOG API key assigned to you if participating in Sogni Supernet debug logs.

COOLDOWN_EVERY_X_JOBS_COUNT=100

By default workers are configured to take a "cool down" break after every 100 successful jobs. You may change this cool down to be at any interval you wish. Note that a value of 0 is considered the default of 100 jobs. We have provided a general default but it is your responsibility to ensure your cooldown values are appropriate for your machine's needs considering some popular models can keep machines active 24/7.

COOLDOWN_EVERY_X_JOBS_SEC=30

During a worker "cool down" this value determines for how many seconds the worker should sit before receiving a new job. The default is 30 seconds. You may change this cool down to be any duration in seconds. Note that a value of 0 will use the default value. During a cooldown a worker will remain connected but just won't receive jobs. There is no penalty to the worker on job priority after leaving cooldown based on cooldown settings. On the contrary, if your cooldown values are too low, and your machine thermally throttles it's performance due to inefficient cooling, that can impact the rate at which it receives jobs, so please cool adequately and set realistic cooldown settings for your scenario.

Setting Up Your docker-compose.yaml File Preferences

• restart: unless-stopped: Remove or comment this line with # if you don't want the worker to start automatically when your computer starts. There are additional options outlined within the docker-compose.yaml file.

• pull_policy: always: This setting ensures Docker Hub checks for the latest version of the worker each time it runs. Comment it out if you want to stay on the current version (not recommended).

• ./data:/data: This maps where Sogni Fast Worker persists configuration. Comment it out if you want all settings to reset with each launch.

• ./data-models:/data-models: This maps where downloaded models are stored between reboots. Comment it out if you don't want to persist models across reboots. You can also manually place models here or map this folder to another location, like an existing Automatic1111 install, to share models.

Running Fast Worker on a machine with multiple GPUs

Running Fast Worker on multiple GPUs is fully supported but the install process is not quite as streamlined as our single GPU installs. Please reach out if you have any difficulty.

Running Fast Worker on multi-GPU machines works by simply running multiple independent copies of the Fast Worker Docker software, each targeting a separate GPU. Model cache and disk space is shared between the instances but each instance will require 30gb of system ram. For example: 3 GPUs = 90gb system ram required. Anything less may present instability and lack of jobs directed at the worker.

1. After downloading the correct Fast Worker installer for your OS the first step is to run the generator script which will create a docker-compose-generated.ymlfile for each of your GPUs. By default the script uses nvidia-smito detect the number of Nvidia GPUs you have and will generate one for each. If you are putting less GPUs online then what you have available, you may just delete the extra ones after running it: ./worker-generator.bat or ./worker-generator.sh.

2. You'll need to open each generated docker-compose-generated.yml file and update the filename listed under "env_file" from the default of ".env" to ".env1" etc. Each GPU will need to point to a different ENV VARS configuration file.

3. In the install folder you will need to duplicate the original .env file supplied to create separate instances that match the names you put in step 2. For example .env1 and .env2. You would then open each with a text editor and populate the required settings, namely API_KEY and NFT_TOKEN_ID. Note that you will need a separate Fast Worker NFT for each worker. So each should have a separate NFT_TOKEN_ID and API_KEY value in each separate .env file. If you are running all workers under the same Sogni account then all NFTs need to be moved to that Sogni account address and the same API_KEY for that address can be used but each worker, however each worker still needs a different NFT_TOKEN_ID that exists in the account tied to that API_KEY. (If you try to run multiple workers using the same NFT_TOKEN_ID only the last one that connects will be able to connect and may be knocked offline by subsequent login attempts by your other workers.)

4. By default each worker runs the Stable-diffusion worker image. If you would like to switch any to the flux worker you can update this line in the generated docker-compose to `sogni/sogni-flux-worker:latest`

   image: sogni/sogni-stable-diffusion-worker:latest

5. Run this command from the installer folder to start all workers:

   docker compose -f docker-compose-generated.yml up -d

6. Run this command from the installer folder to stop all workers:

   docker compose -f docker-compose-generated.yml down

7. To view the health status of all running docker workers

   docker compose ps

8. To tail the logs of a particular worker instance for troubleshooting grab the ID from docker compose psand run the following:

   docker logs -f ID_OF_DOCKER_INSTANCE_FROM_DOCKER_PS

Need help? Join our Discord!

If you have any issues or questions setting up your Sogni Fast Worker you can reach out on Discord or via email to [e-mail] for technical support. Join Sogni Discord ✨