* This page is part of the Union.ai v2 docs.
* Full index: https://www.union.ai/docs/v2/union/llms.txt
* Union.ai is a commercial superset of the open-source Flyte project and uses the same `flyte` SDK and CLI, so questions about Flyte are usually answered here too.
* Only refer to the open-source Flyte docs if you are explicitly looking for information about the open-source Flyte project, particularly the open-source Flyte backend. The open-source Flyte v2 docs are at https://www.union.ai/docs/v2/flyte/llms.txt
* These are the v2 docs. v1 is out of date and deprecated. Only refer to v1 docs if asked explicitly about v1. The v1 index is at https://www.union.ai/docs/v1/union/llms.txt

Naming, metadata, default inputs, and environment variables, and the settings without their own page.

# Additional task settings

This page covers task configuration parameters that do not have their own dedicated page:
naming and metadata, default inputs, environment variables, and inline I/O thresholds.

For the full list of all task configuration parameters, see [Configure tasks](https://www.union.ai/docs/v2/union/user-guide/tasks/task-configuration.md).

## Naming and metadata

### `name`

The `name` parameter on `TaskEnvironment` is required.
It is combined with each task function name to form the fully-qualified task name.
For example, if you define a `TaskEnvironment` with `name="my_env"` and a task function `my_task`,
the fully-qualified task name is `my_env.my_task`.

The `name` must use `snake_case` or `kebab-case` and is immutable once set.

### `short_name`

The `short_name` parameter on `@env.task` (and `override()`) overrides the display name of a task in the UI graph view.
By default, the display name is the Python function name.
Overriding `short_name` does not change the fully-qualified task name.

```python
import flyte

env = flyte.TaskEnvironment(name="my_env")

@env.task(short_name="Train Model")
def train(data: list) -> dict:
    return {"accuracy": 0.95}
```

### `description`

The `description` parameter on `TaskEnvironment` provides a description of the task environment (max 255 characters).
It is used for organizational purposes and can be viewed in the UI.

### `docs`

A task's description, shown in the UI, comes from the task function's docstring:

```python
import flyte

env = flyte.TaskEnvironment(name="my_env")

@env.task
def train(data: list) -> dict:
    """Trains a model on the given dataset."""
    return {"accuracy": 0.95}
```

> [!NOTE]
> `@env.task` also accepts a `docs` parameter. In the current SDK it sets only the help text that
> `flyte run` shows for the task; it does not change the description shown in the UI. Use the docstring.

### `report`

The `report` parameter on `@env.task` controls whether an HTML report is generated for the task.
See [Reports](https://www.union.ai/docs/v2/union/user-guide/tasks/task-programming/reports.md) for details.

### Source-code link (automatic)

In addition to the description and docs above, Flyte automatically attaches a link from each deployed task back to its source code on GitHub or GitLab, when you deploy from inside a checked-out git repository.
The link is rendered next to the task description in the UI.
There is no parameter to set. See [Source-code link discovery](https://www.union.ai/docs/v2/union/user-guide/tasks/task-deployment/how-task-deployment-works.md#6-source-code-link-discovery) for the conditions and caveats.

### `links`

The `links` parameter on `@env.task` (and `override()`) attaches clickable URLs to tasks in the UI.
Use links to connect tasks to external tools like experiment trackers, monitoring dashboards, or logging systems.

Links are defined by implementing the [`Link`](https://www.union.ai/docs/v2/union/api-reference/flyte-sdk/flyte/link.md) protocol.
See [Links](https://www.union.ai/docs/v2/union/user-guide/tasks/task-programming/links.md) for full details on creating and using links.

## Default inputs

Task functions support Python default parameter values. When a task parameter has a default, callers can omit it and the default is used.

```python
import flyte

env = flyte.TaskEnvironment(name="my_env")

@env.task
async def process(data: list, batch_size: int = 32, verbose: bool = False) -> dict:
    # batch_size defaults to 32, verbose defaults to False
    ...
```

When running via `flyte run`, parameters with defaults are optional:

```bash
# Uses defaults for batch_size and verbose
flyte run my_file.py process --data '[1, 2, 3]'

# Override a default
flyte run my_file.py process --data '[1, 2, 3]' --batch-size 64
```

When invoking programmatically, Python's normal default argument rules apply:

```python
result = flyte.run(process, data=[1, 2, 3])          # batch_size=32, verbose=False
result = flyte.run(process, data=[1, 2, 3], batch_size=64)  # override
```

Defaults are part of the task's input schema and are visible in the UI when viewing the task.

## Environment variables

The `env_vars` parameter on `TaskEnvironment` injects plain-text environment variables into the task container.
It accepts a `Dict[str, str]`.

```python
import flyte

env = flyte.TaskEnvironment(
    name="my_env",
    env_vars={
        "LOG_LEVEL": "DEBUG",
        "API_ENDPOINT": "https://api.example.com",
    },
)

@env.task
def my_task() -> str:
    import os
    return os.environ["API_ENDPOINT"]
```

Environment variables can be overridden at the `task.override()` invocation level
(unless `reusable` is in effect).

Use `env_vars` for non-sensitive configuration values.
For sensitive values like API keys and credentials, use [`secrets`](https://www.union.ai/docs/v2/union/user-guide/tasks/task-configuration/secrets.md) instead.

## Inline I/O threshold

The `max_inline_io_bytes` parameter on `@env.task` (and `override()`) controls the maximum
size for data passed directly in the task request and response
(e.g., primitives, strings, dictionaries).

Data exceeding this threshold raises an `InlineIOMaxBytesBreached` error.

The default value is 10 MiB (`10 * 1024 * 1024` bytes).

This setting does **not** affect [`flyte.io.File`, `flyte.io.Dir`](https://www.union.ai/docs/v2/union/user-guide/tasks/task-programming/files-and-directories.md),
or [`flyte.io.DataFrame`](https://www.union.ai/docs/v2/union/user-guide/tasks/task-programming/dataframes.md),
which are always offloaded to object storage regardless of size.

```python
import flyte

env = flyte.TaskEnvironment(name="my_env")

# Allow up to 50 MiB of inline data
@env.task(max_inline_io_bytes=50 * 1024 * 1024)
def process_large_dict(data: dict) -> dict:
    return {k: v * 2 for k, v in data.items()}
```

---
**Source**: https://github.com/unionai/unionai-docs/blob/main/content/user-guide/tasks/task-configuration/additional-task-settings.md
**HTML**: https://www.union.ai/docs/v2/union/user-guide/tasks/task-configuration/additional-task-settings/