synth-subnet

Miner Reference

Table of contents

• 1. Options
  • 1.1. Common Options
    • --axon.port INTEGER
    • --blacklist.allow_non_registered BOOLEAN
    • --blacklist.force_validator_permit BOOLEAN
    • --blacklist.validator_min_stake INTEGER
    • --logging.debug
    • --logging.info
    • --logging.trace
    • --netuid INTEGER
    • --neuron.device TEXT
    • --neuron.dont_save_events BOOLEAN
    • --neuron.epoch_length INTEGER
    • --neuron.events_retention_size TEXT
    • --neuron.name TEXT
    • --neuron.timeout INTEGER
    • --neuron.vpermit_tao_limit INTEGER
    • --wallet.hotkey TEXT
    • --wallet.name TEXT
  • 1.2. Weights & Bases Options
    • --wandb.enabled BOOLEAN
    • --wandb.entity TEXT
    • --wandb.project_name TEXT
• 2. Appendix
  • 2.1. FAQs
  • 2.2. Useful Commands
  • 2.3. Troubleshooting

1. Options

1.1. Common Options

--axon.port INTEGER

The external port for the Axon component. This port is used to communicate to other neurons.

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--axon.port 8091",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --axon.port 8091

^{Back to top ^}

--blacklist.allow_non_registered BOOLEAN

If set, miners will accept queries from non-registered entities.

🚨 WARNING: Make sure you know what you are doing when setting this option.

Default: false

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--blacklist.allow_non_registered true",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --blacklist.allow_non_registered true

^{Back to top ^}

--blacklist.force_validator_permit BOOLEAN

If set, we will force incoming requests to have a permit.

Default: false

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--blacklist.force_validator_permit true",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --blacklist.force_validator_permit true

^{Back to top ^}

--blacklist.validator_min_stake INTEGER

Minimum validator stake to accept forward requests from as a miner, (e.g. 1000).

Default: 1000

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--blacklist.validator_min_stake 1000",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --blacklist.validator_min_stake 1000

^{Back to top ^}

--logging.debug

Turn on bittensor debugging information.

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--logging.debug",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --logging.debug

^{Back to top ^}

--logging.info

Turn on bittensor info level information.

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--logging.info",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --logging.info

^{Back to top ^}

--logging.trace

Turn on bittensor trace level information.

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--logging.trace",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --logging.trace

^{Back to top ^}

--netuid INTEGER

The netuid (network unique identifier) of the subnet within the root network, (e.g. 247).

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--netuid 247",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --netuid 247

^{Back to top ^}

--neuron.device TEXT

The name of the device to run on.

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--neuron.device cuda",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --neuron.device cuda

^{Back to top ^}

--neuron.dont_save_events BOOLEAN

Whether events are saved to a log file.

Default: false

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--neuron.dont_save_events true",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --neuron.dont_save_events true

^{Back to top ^}

--neuron.epoch_length INTEGER

The default epoch length (how often we set weights, measured in 12 second blocks), (e.g. 100).

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--neuron.epoch_length 100",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --neuron.epoch_length 100

^{Back to top ^}

--neuron.events_retention_size TEXT

The events retention size.

Default: 2147483648 (2GB)

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--neuron.events_retention_size 2147483648",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --neuron.events_retention_size 2147483648

^{Back to top ^}

--neuron.name TEXT

Trials for this neuron go in neuron.root / (wallet_cold - wallet_hot) / neuron.name.

Default: miner

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--neuron.name miner",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --neuron.name miner

^{Back to top ^}

--neuron.timeout INTEGER

The maximum timeout in seconds of the miner neuron response, (e.g. 120).

Default: -

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--neuron.timeout 120",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --neuron.timeout 120

^{Back to top ^}

--neuron.vpermit_tao_limit INTEGER

The maximum number of TAO allowed that is allowed for the validator to process miner response, (e.g. 1000).

Default: 4096

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--neuron.vpermit_tao_limit 1000",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --neuron.vpermit_tao_limit 1000

^{Back to top ^}

--wallet.hotkey TEXT

The hotkey of the wallet.

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--wallet.hotkey default",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --wallet.hotkey default

^{Back to top ^}

--wallet.name TEXT

The name of the wallet to use.

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--wallet.name miner",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --wallet.name miner

^{Back to top ^}

1.2. Weights & Bases Options

--wandb.enabled BOOLEAN

Boolean toggle for wandb integration.

Default: false

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--wandb.enabled true",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --wandb.enabled true

^{Back to top ^}

--wandb.entity TEXT

The username or team name where you want to send W&B runs.

Default: opentensor-dev

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--wandb.entity opentensor-dev",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --wandb.entity opentensor-dev

^{Back to top ^}

--wandb.project_name TEXT

The name of the project where W&B runs.

Default: template-miners

Example:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--wandb.project_name template-miners",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

Alternatively, you can add the args directly to the command:

pm2 start miner.config.js -- --wandb.project_name template-miners

^{Back to top ^}

2. Appendix

2.1. FAQs

1. What is the best way for me to track my miner’s performance?

Visit the Synth Miner Dashboard: https://miners.synthdata.co.

2. How long do I have to wait for my new miner to get the first CRPS score?

Assuming your setup is correct, and you’re submitting predictions, it takes approximately 25–27 hours for your first score to appear. This is because the CRPS is calculated based on the actual price 24 hours after the prediction.

3. What does it mean when I check https://api.synthdata.co/validation/miner?uid= and see: {"validated":false,"reason":"Number of time points is incorrect: expected 289, got 288","response_time":"4.07"}?

It means your submission has the wrong number of time points. You should be submitting exactly 289 points for that prompt.

4. Why does my miner keep getting a CRPS score of -1?

Your miner will receive a CRPS score of -1 if the prediction is not in the correct format.

5. What is the correct format for subnet-prompted predictions?

A prediction is considered valid if it meets all the following conditions:

• Submitted before the timeout specified by the validator.
• Contains the required number of simulation paths (num_simulations in the prompt).
• Each path has the correct number of time points:

expected_time_points = (time_length / time_increment) + 1

The expected format is as follows:

[
  start_timestamp, time_interval,
  [path1_price_t0, path1_price_t1, path1_price_t2, ...], [path2_price_t0, path2_price_t1, path2_price_t2, ...],
  ...
]

An example of a valid response would be:

[
  1760084861, 300,
  [104856.23, 104972.01, 105354.9, ...], [104856.23, 104724.54, 104886, ...], [104856.23, 104900.12, 104950.45, ...]
  ...
]

where

• the first element is the timestamp of the start time of the prompt,
• second is the time increment of the prompt,
• then arrays of prices with no more than 8 digits per point. Otherwise, the validator will reject the submission.
  • validation error message: Price format is incorrect: too many digits

You can find the validation function here:

response_validation_v2.py#L51

And an example of the prompt parameters here:

prompt_config.py#L5

You can test your prediction format using this miner script:

python synth/miner/run.py

Expected output:

start_time 2025-11-10T13:29:00+00:00
CORRECT

6. Synth is predicting multiple assets. How does each asset prediction contribute to the smoothed score?

Each asset has its own prompts and is scored separately: for every asset/prompt pair, the validator computes a CRPS-based prompt_score_v3 over time. These per-asset scores are then combined into a single smoothed score using a time-based moving average with asset-specific coefficients (weights). Assets with higher coefficients contribute more to the final smoothed score than lower-weighted assets.

7. I am getting fair CRPS scores but my reward weight is still zero. Why?

Miner reward weights are determined by a softmax function over a smoothed average of your CRPS scores over the past 10 days, using a 5-day half-life. This is subject to future change.

8. How do I improve my miner performance?

Synth regularly publishes detailed miner performance reviews highlighting the strategies of top-performing miners. These reviews can help you optimize your own models.

Explore them here: https://mirror.xyz/synthdata.eth

9. Do I need to run multiple miners?

If you’re using the same model per asset, no — rewards will be the same. If you’re using different models, then yes, it’s encouraged to experiment and find what works best.

10. Is there a penalty for turning off my miner for a period of time?

If you miss a prompt or submit in an incorrect format, your prompt score will be penalized—set to the 90th percentile of all prompt scores, and your CRPS will be -1.

If you make the best prediction, your prompt score will be 0, and your CRPS will also be at its minimum of 0, meaning your prediction perfectly matched the realized price.

The smoothed score is calculated as the moving average of your prompt scores over time.

These rules apply across all assets.

11. Could someone explain how trust values are calculated?

Trust is computed based on the subnet’s scoring system, outlined here: https://docs.bittensor.com/emissions#trust

It aggregates individual miner scores into a consensus-based weighting system, which determines reward distribution based on prediction quality.

12. How much time do miners have to respond to a prompt?

Each prompt contains a start_time, which acts as the deadline for your response. You’ll always have at least 2 minutes from when the prompt is sent. We recommend submitting your response within 40 seconds.

13. Is Synth running on a testnet too?

Yes. The testnet UID is 247, and it functions identically to mainnet.

14. Once I start a miner using the guide on GitHub, where do I change the code to run my own model?

Modify this function:

simulations.py#L25

Take into account all prompt parameters except sigma, which you may ignore.

^{Back to top ^}

2.2. Useful Commands

Command	Description
pm2 stop miner	Stops the miner.
pm2 logs miner --lines 100	View the logs of the miner.

^{Back to top ^}

2.3. Troubleshooting

ModuleNotFoundError: No module named 'simulation'

This error is due to Python unable to locate the local Python modules. To avoid this error, ensure you have created and activate the Python virtual environment from the project root and ensure the PYTHONPATH is present in the <miner|validator>.config.js file and is pointing to the project root.

An example of a config file should be:

// miner.config.js
module.exports = {
  apps: [
    {
      name: "miner",
      interpreter: "python3",
      script: "./neurons/miner.py",
      args: "--netuid 247 --logging.debug --logging.trace --subtensor.network test --wallet.name miner --wallet.hotkey default --axon.port 8091",
      env: {
        PYTHONPATH: ".",
      },
    },
  ],
};

As you can see, we are setting the PYTHONPATH environment variable that will be injected when we run pm2 start miner.

^{Back to top ^}

This site is open source. Improve this page.