Loading
Loading
This information was auto-extracted from the GitHub repository and may be incomplete or out of date. Always refer to the subnet's GitHub repository for the latest instructions.
GPU hardware only needed for local testing. Docker image must not exceed 20 GB (model weights are mounted at runtime).
Two components: always-on CPU host for chain scanning and weight setting; ephemeral GPU pod rented on-demand for evaluation when challengers are queued. GPU pod requires 8-GPU NVLink/SXM configuration.
Tag your inference server Docker image with public registry path
docker tag my-server:latest docker.io/myuser/cacheon-miner:v1
Push image to public container registry
docker push docker.io/myuser/cacheon-miner:v1
Pay submission fee and commit image reference on-chain
python miner/commit.py --wallet-name <wallet> --wallet-hotkey <hotkey> --image 'docker.io/myuser/cacheon-miner:v1' --digest 'sha256:...' --fee 0.1 --network finney --netuid 14
Clone Cacheon repository
git clone https://github.com/latent-to/cacheon && cd cacheon
Create environment file and add wallet and S3 configuration
cp .env.example .env
Start always-on CPU host component for chain scanning and weight setting
docker compose up --build
Run GPU pod provisioning script (Targon or Lium) with API credentials and customized parameters(optional)
bash scripts/gpu_setup/setup.sh
Start GPU pod component for on-demand evaluation (run when challengers appear)(optional)
docker compose -f validator/gpu-compose.yml up --build -d