> For the complete documentation index, see [llms.txt](https://docs.tensorusd.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.tensorusd.com/user-type/miners.md). # Miners ### What Miners Do Miners participate in liquidation auctions on the TensorUSD protocol. A vault is liquidated when its collateral value drops below 1.2× the debt (liquidation ratio). At that point the vault is put up for auction. Miners bid TUSDT to win the vault's collateral. Because the collateral is worth at least 1.2× the debt at liquidation, a miner who bids close to the debt amount captures the spread as profit. On top of that, winning auctions earns the miner TAO emissions from the Bittensor subnet. *** ### How Miners Earn There are two sources of income: #### 1. On-chain Profit (from the auction itself) The miner pays the winning bid in TUSDT and receives the vault's collateral tokens. `liquidation triggers when:` `collateral_value < liquidation_ratio × debt` `liquidation_ratio = 1.2` `profit = collateral_value - winning_bid` `where:` `collateral_value = collateral_amount × collateral_price (in TUSDT)` `winning_bid = amount paid by the miner (in TUSDT)` Because the liquidation ratio is 1.2, `collateral_value ≥ 1.2 × debt` is guaranteed at the point of liquidation. This creates a built-in spread of at least 20% above debt for miners to capture. Example: * Vault collateral: `1 TAO` at price `150 TUSDT → collateral_value = 150 TUSDT` * Vault debt: `100 TUSDT` * Liquidation ratio: `1.2` → vault goes to auction when collateral value falls below `120 TUSDT` * Collateral value at auction: `120 TUSDT` (1.2 × 100) * Miner's winning bid: `110 TUSDT` * On-chain profit: `120 - 110 = 10 TUSDT` #### 2. TAO Emissions (from Bittensor subnet rewards) Every epoch, the validator reads all auction wins from that period and assigns weights to miners. Those weights determine each miner's share of the subnet's TAO emissions. *** ### Incentive Mechanism (Reward Formula) The validator scores each miner's win using a linear reward function based on how much the winning bid exceeds the vault's debt: `bonus_ratio = min((winning_bid - debt_balance) / debt_balance, 0.20)` `reward = 1.0 + bonus_ratio` | Winning Bid vs Debt | bonus\_ratio | reward | | --------------------------- | ------------ | ------ | | Exactly debt `(bid = debt)` | `0.00` | `1.0` | | 10% over debt | `0.10` | `1.1` | | 20% over debt | `0.20` | `1.2` | | 30% over debt (capped) | `0.20` | `1.2` | * Minimum reward per win: **1.0** * Maximum reward per win: **1.2** * The bonus caps at **20% overpay** — bidding more than that gives no extra TAO reward. **Multiple wins in the same epoch accumulate:** `total_reward(miner) = sum of reward(win_i) for all wins in tempo window` So a miner who wins three auctions in one epoch with rewards 1.0 + 1.5 + 2.0 gets a total score of 4.5 before normalisation. #### Weight Setting At the end of each epoch the validator: 1. Collects all per-miner accumulated rewards. 2. Applies an exponential moving average (EMA) to smooth scores over time: `score = α × new_reward + (1 - α) × previous_score` 3. L1-normalises all scores across miners and submits them as weights to the chain. Each miner's TAO emission share is proportional to their normalised weight. *** ### Bidding Strategy The miner uses a configurable strategy to stay profitable: `initial_bid_percentage = 0.05 → initial bid = debt × 1.05` `bid_increment_rate = 0.05 → rebid = current_highest × 1.05` `max_bid_percentage = 0.95 → max bid = debt × 1.95` `max_bid_absolute = None → no absolute token cap (optional override)` `min_profit_margin = 0.02 → reject bid if profit < 2% of collateral_value` Since the liquidation ratio is 1.2, the collateral is worth at least 1.2 × debt at auction time. The theoretical maximum spread available to a miner is therefore: `max possible profit = collateral_value - debt ≥ (1.2 × debt) - debt = 0.2 × debt` max\_bid\_percentage = 0.95 is a ceiling. min\_profit\_margin = 0.02 is a profitability guard. The bid is rejected before submission if: `collateral_value - bid < collateral_value × 0.02` The miner automatically re-evaluates and counter-bids whenever it is outbid, up to the configured cap. *** ### How to Run `uv run neurons/miner.py \` `--netuid 421 \` `--subtensor.network test \` `--wallet.name \` `--wallet.hotkey \` `--logging.info`
| Flag | Description | | -------------------------- | ---------------------------- | | `--netuid 421` | Subnet UID to register on | | `--subtensor.network test` | Connect to the testnet | | `--wallet.name test` | Name of the local wallet | | `--wallet.hotkey default` | Hotkey to use for this miner | | `--logging.info` | Enable info-level logs | --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.tensorusd.com/user-type/miners.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.