> For the complete documentation index, see [llms.txt](https://docs.stacks.co/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.stacks.co/operate/run-a-miner/miner-costs-and-fees.md).

# Miner Costs and Fees

### Configuring Cost and Fee Estimation

{% code title="config.toml" %}

```toml
[fee_estimation]
cost_estimator = naive_pessimistic
fee_estimator = scalar_fee_rate
fee_rate_fuzzer_fraction = 0.1
fee_rate_window_size = 5
cost_metric = proportion_dot_product
log_error = false
enabled = true
```

{% endcode %}

Fee and cost estimators observe network transactions and build estimates for viable fee rates and expected execution costs from the costs they see. `log_error` controls whether the INFO logger will display information about cost estimator accuracy as new costs are observed. Setting `enabled = false` turns off the cost estimators.

Accepted values, as of stacks-core `4.0.3`:

| Key              | Accepted values                                      | Default                  |
| ---------------- | ---------------------------------------------------- | ------------------------ |
| `cost_estimator` | `naive_pessimistic`                                  | `naive_pessimistic`      |
| `fee_estimator`  | `scalar_fee_rate`, `fuzzed_weighted_median_fee_rate` | `scalar_fee_rate`        |
| `cost_metric`    | `proportion_dot_product`                             | `proportion_dot_product` |

The sample above is the default configuration. The node lowercases each value and panics at startup on anything else.

{% hint style="info" %}
Cost estimators are not consensus-critical components: they are intended for miners to rank mempool transactions or for clients to pick appropriate fee rates before broadcasting.
{% endhint %}

The `fuzzed_weighted_median_fee_rate` estimator:

* uses a median estimate from a window of the fees paid in the last `fee_rate_window_size` blocks, and
* then applies a uniform random "fuzz" up to `fee_rate_fuzzer_fraction` of the base estimate.

### The Epoch 4.0 block read budget

Epoch 4.0 doubles the read fields of the Clarity block execution budget. Write and runtime limits are unchanged.

| Field          | Epoch 2.1 through 3.4 | Epoch 4.0     |
| -------------- | --------------------- | ------------- |
| `read_length`  | 100,000,000           | 200,000,000   |
| `read_count`   | 15,000                | 30,000        |
| `write_length` | 15,000,000            | 15,000,000    |
| `write_count`  | 15,000                | 15,000        |
| `runtime`      | 5,000,000,000         | 5,000,000,000 |

This is the per-block budget, not the per-function cost table. Individual Clarity function costs are unchanged, so a block reaches its read limit at twice the previous volume. The config keys and estimator names above are unaffected.

Read the limits your node is enforcing:

```bash
curl -s http://localhost:20443/v2/pox \
  | jq '.epochs[] | select(.epoch_id=="Epoch40") | .block_limit'
```

<details>

<summary>Mining calculator (external)</summary>

Mining calculator: <https://friedger.github.io/mining-calculator/>

Source: <https://github.com/friedger/mining-calculator>

</details>


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.stacks.co/operate/run-a-miner/miner-costs-and-fees.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.