Prerequisites

Before setting up your miner, verify your hardware and software environment meets these requirements.

Hardware Requirements

Component	Minimum	Recommended
GPU VRAM	48GB (A6000)	48GB (A6000)
System RAM	32GB	64GB
Storage	100GB SSD	500GB NVMe SSD
Network	Stable broadband	Dedicated server connection

GPU compatibility: NVIDIA A6000 (48GB) is the reference hardware. Other GPUs with 48GB+ VRAM will also work.

Software Requirements

Python 3.10 or later — earlier versions are not supported
CUDA 12.x — must be compatible with your GPU driver
Git — for cloning the repository
pip — for installing dependencies

Verify your CUDA version:

nvidia-smi
nvcc --version

Bittensor Wallet Setup

You need a Bittensor wallet with a coldkey and hotkey to participate in the subnet. Detailed information on managing keys can be found in the official Bittensor guide.

Create a new wallet (Bittensor is installed as part of the project dependencies — no separate install needed):

btcli wallet new_coldkey --wallet.name my-wallet
btcli wallet new_hotkey --wallet.name my-wallet --wallet.hotkey my-hotkey

Check your wallet:

btcli wallet list

TAO Registration

You must register your hotkey on the Connito subnet to participate. Registration requires a small amount of TAO (the Bittensor network token).

btcli subnet register --wallet.name my-wallet --wallet.hotkey my-hotkey --netuid 102

Check the current registration cost (look for "Registration cost (recycled)"):

btcli subnets metagraph --netuid 102

Registration costs TAO. Check the current recycle cost before registering. Registration is permanent for the lifetime of your hotkey on this subnet.

Hugging Face Setup

You must have a Hugging Face account and grant access to the required dataset.

Log in at Hugging Face.
Accept the NVIDIA Open Data License Agreement to the nvidia/Nemotron-CC-Math-v1 dataset.
Install the Hugging Face CLI:
```
pip install -U huggingface_hub
```
Generate a Read only token in your Hugging Face Settings.
Create a .env file in the project root (if it doesn't exist) and add your token:
```
HF_TOKEN=hf_xxxxxxxxxxxxxxxxxxxx
```
Log in using the CLI:
```
huggingface-cli login
```

Checklist

Before proceeding to Installation:

GPU with ≥48GB VRAM confirmed (A6000 or equivalent)
Python 3.10+ installed
CUDA 12.x installed and working (nvidia-smi shows GPU)
Hugging Face account set up with dataset access and token in .env
Bittensor wallet created (coldkey + hotkey)
TAO obtained for registration
Subnet registration completed
Optional: Telemetry stack ready (Grafana & Prometheus)