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This information was auto-extracted from the GitHub repository and may be incomplete or out of date.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 with minimum 6.0 CUDA compute capability required (7.0 recommended). SSD storage preferred with minimum 1000 IOPS (5000 IOPS recommended). CPU-only variant available (gpu_required: false) for miners using OpenAI API only.
GPU with minimum 6.0 CUDA compute capability required (7.0 recommended). SSD storage preferred with minimum 1000 IOPS (5000 IOPS recommended). Validators require COMMITMENT_PRIVATE_KEY environment variable to decrypt miner endpoints and access ReadyAI Conversation Server API.
Clone the ReadyAI subnet repository
git clone https://github.com/afterpartyai/bittensor-conversation-genome-project.git cgp-subnet && cd cgp-subnet
Install Python dependencies including bittensor, torch, openai, and other required packages
pip install -r requirements.txt
Create environment configuration file from template
cp env.example .env
Edit .env to add OpenAI API key, Weights & Biases key, and configure LLM settings (GPT-4o is default)
nano .env (or your preferred editor)
Run miner loop tests to verify configuration is correct before registering on network(optional)
python -m pytest -s --disable-warnings tests/test_full_loop.py
Register your hotkey on the subnet (testnet or mainnet)
btcli s register --netuid 138 --wallet.name <coldkey_name> --wallet.hotkey <hotkey_name> --network test --wallet.path ~/.bittensor/wallets (or --netuid 33 --network finney for mainnet)
Start the miner (use --netuid 33 without --subtensor.network for mainnet)
python3 -m neurons.miner --subtensor.network test --netuid 138 --wallet.name <coldkey_name> --wallet.hotkey <hotkey_name> --logging.debug --axon.port <port>
Clone the ReadyAI subnet repository
git clone https://github.com/afterpartyai/bittensor-conversation-genome-project.git cgp-subnet && cd cgp-subnet
Install Python dependencies including bittensor, torch, openai, wandb, and other required packages
pip install -r requirements.txt
Create environment configuration file from template
cp env.example .env
Edit .env to add OpenAI API key, Weights & Biases key (required for validators), COMMITMENT_PRIVATE_KEY, and configure LLM settings
nano .env (or your preferred editor)
Generate ReadyAI Conversation Server API key by running the validator API key generation script. Requires wallet credentials and will create readyai_api_data.json file
python scripts/get_validator_api_key.py (or 'test' for testnet)
Register your hotkey on the subnet with sufficient stake (minimum stake required for validator permit)
btcli s register --netuid 138 --wallet.name <coldkey_name> --wallet.hotkey <hotkey_name> --network test --wallet.path ~/.bittensor/wallets (or --netuid 33 --network finney for mainnet)
Start the validator (use --netuid 33 without --subtensor.network for mainnet)
python3 -m neurons.validator --subtensor.network test --netuid 138 --wallet.name <coldkey_name> --wallet.hotkey <hotkey_name> --logging.debug --axon.port <port>
Set the COMMITMENT_PRIVATE_KEY environment variable (testnet key provided in env.example; mainnet key distributed securely to validators)
export COMMITMENT_PRIVATE_KEY=<provided_key>