Run a watchtower!

You can deploy one easily using Docker!

🚀 Option 1: With NVIDIA GPU (Recommended)

For the best performance and support for larger models, run your Watchtower using a CUDA-enabled NVIDIA GPU:

docker run \
 -d \
 --gpus all \
 --network=host \
 -e WALLET_PUBLIC_KEY=0x_your_key_here \
 -e MODEL_NAME=MODEL_NAME \
 -e NETWORK=testnet \
 --name wtns-test \
 wtns-ai.tail1d7c2.ts.net:5000/infinitywatch-nvidia:dev

🧪 Option 2: CPU-Only (Lightweight Model)

No GPU? You can still join the network by running a smaller model on your CPU:

docker run \
 -d \
 -e WALLET_PUBLIC_KEY=0x_your_key_here \
 -e MODEL_NAME=MODEL_NAME \
 -e NETWORK=testnet \
 --name wtns-test \
 wtns-ai.tail1d7c2.ts.net:5000/infinitywatch:dev

Note:

following are the models we support as of now;

🤗 HuggingFace family
HuggingFaceTB/SmolVLM2-2.2B-Instruct
HuggingFaceTB/SmolVLM-500M-Instruct
HuggingFaceTB/SmolVLM-256M-Instruct

🔮 Qwen Family
Qwen/Qwen2.5-VL-3B-Instruct
Qwen/Qwen2.5-VL-7B-Instruct