Run Config

RunConfig controls dataset generation behavior, including early shutdown thresholds, batch sizing, non-inference worker concurrency, and the Jinja rendering engine used by the runtime.

JinjaRenderingEngine.SECURE is the default. Set JinjaRenderingEngine.NATIVE when you want Jinja2's broader built-in sandbox behavior instead of Data Designer's hardened renderer.

For guidance on when to use each mode, see Security.

Usage

import data_designer.config as dd
from data_designer.interface import DataDesigner

data_designer = DataDesigner()
data_designer.set_run_config(dd.RunConfig(
    buffer_size=500,
    max_conversation_restarts=3,
    jinja_rendering_engine=dd.JinjaRenderingEngine.NATIVE,
))

API Reference

Classes:

Name	Description
`JinjaRenderingEngine`	Template renderer used by the engine for user-supplied Jinja templates.
`RequestAdmissionTuningConfig`	Advanced request-admission AIMD tuning for model API calls.
`RunConfig`	Runtime configuration for dataset generation.
`ThrottleConfig`	Deprecated compatibility DTO for request-admission tuning.

`JinjaRenderingEngine`

Bases: StrEnum

Template renderer used by the engine for user-supplied Jinja templates.

`RequestAdmissionTuningConfig`

Bases: ConfigBase

Advanced request-admission AIMD tuning for model API calls.

Most workloads should tune model capacity with max_parallel_requests on inference parameters. These fields adjust the adaptive recovery behavior below that cap and are intended for provider/runtime support cases.

`RunConfig`

Bases: ConfigBase

Runtime configuration for dataset generation.

Groups configuration options that control generation behavior but aren't part of the dataset configuration itself.

Attributes:

Name	Type	Description
`disable_early_shutdown`	`bool`	If True, disables the executor's early-shutdown behavior entirely. Generation will continue regardless of error rate, and the early-shutdown exception will never be raised. Error counts and summaries are still collected. Default is False.
`shutdown_error_rate`	`float`	Error rate threshold (0.0-1.0) that triggers early shutdown when early shutdown is enabled. Default is 0.5.
`shutdown_error_window`	`int`	Minimum number of completed tasks before error rate monitoring begins. Must be >= 1. Default is 10.
`buffer_size`	`int`	Number of records to process in each batch during dataset generation. A batch is processed end-to-end (column generation, post-batch processors, and writing the batch to artifact storage) before moving on to the next batch. Must be > 0. Default is 1000.
`non_inference_max_parallel_workers`	`int`	Maximum number of worker threads used for non-inference cell-by-cell generators. Must be >= 1. Default is 4.
`max_conversation_restarts`	`int`	Maximum number of full conversation restarts permitted when generation tasks call `ModelFacade.generate(...)`. Must be >= 0. Default is 5.
`max_conversation_correction_steps`	`int`	Maximum number of correction rounds permitted within a single conversation when generation tasks call `ModelFacade.generate(...)`. Must be >= 0. Default is 0.
`async_trace`	`bool`	If True, collect per-task tracing data when using the async engine (DATA_DESIGNER_ASYNC_ENGINE=1). Has no effect on the sync path. Default is False.
`progress_bar`	`bool`	If True, display sticky ANSI progress bars instead of periodic log lines during generation. Requires a TTY; falls back to log lines in non-TTY environments. Default is False.
`progress_interval`	`float`	How often (in seconds) the async progress reporter emits a consolidated log block. Must be > 0. Default is 5.0.
`jinja_rendering_engine`	`JinjaRenderingEngine`	Template renderer used for engine-side Jinja evaluation. `native` uses Jinja2's built-in sandbox with the standard filter set and fewer Data Designer-specific restrictions. `secure` uses Data Designer's hardened sandbox with additional AST, filter, and output guards. Default is `secure`.
`request_admission`	`RequestAdmissionTuningConfig \| None`	Advanced AIMD request-admission tuning for provider/model calls. Most users should leave this unset and tune `max_parallel_requests` instead.

Notes

Request-admission controller internals remain engine-owned. This field exposes only the supported tuning DTO and does not expose controller mutation APIs, leases, queues, or pressure snapshots.

Methods:

Name	Description
`normalize_shutdown_settings`	Normalize shutdown settings for compatibility.

`normalize_shutdown_settings()`

Normalize shutdown settings for compatibility.

Source code in packages/data-designer-config/src/data_designer/config/run_config.py

@model_validator(mode="after")
def normalize_shutdown_settings(self) -> Self:
    """Normalize shutdown settings for compatibility."""
    if self.disable_early_shutdown:
        self.shutdown_error_rate = 1.0
    return self

`ThrottleConfig`

Bases: ConfigBase

Deprecated compatibility DTO for request-admission tuning.

Use RequestAdmissionTuningConfig via RunConfig.request_admission instead. ceiling_overshoot is accepted for compatibility but is not forwarded because request admission no longer exposes an overshoot knob.

Methods:

Name	Description
`to_request_admission_tuning`	Translate legacy throttle tuning into the request-admission DTO.

`to_request_admission_tuning()`

Translate legacy throttle tuning into the request-admission DTO.

Source code in packages/data-designer-config/src/data_designer/config/run_config.py

def to_request_admission_tuning(self) -> RequestAdmissionTuningConfig:
    """Translate legacy throttle tuning into the request-admission DTO."""
    return RequestAdmissionTuningConfig(
        multiplicative_decrease_factor=self.reduce_factor,
        additive_increase_step=self.additive_increase,
        successes_until_increase=self.success_window,
        cooldown_seconds=self.cooldown_seconds,
        startup_ramp_seconds=self.rampup_seconds,
    )