dragon.ai.inference.config.InferenceConfig

class InferenceConfig

Bases: object

Master configuration for the entire inference pipeline.

Only model is required. All other fields have sensible defaults suitable for agentic pipelines.

__init__(model: ModelConfig, hardware: HardwareConfig = <factory>, batching: BatchingConfig = <factory>, guardrails: GuardrailsConfig = <factory>, dynamic_worker: DynamicWorkerConfig = <factory>, flask_secret_key: str = '', run_type: str = 'chat', token: str = '') → None

Methods

__init__(model, hardware, batching, ...)
from_dict(config_dict)	Create InferenceConfig from dictionary (loaded from YAML).
validate_all(all_nodes)	Validate all configuration sections.

Attributes

flask_secret_key
run_type
token
model
hardware
batching
guardrails
dynamic_worker

model: ModelConfig

hardware: HardwareConfig

batching: BatchingConfig

guardrails: GuardrailsConfig

dynamic_worker: DynamicWorkerConfig

flask_secret_key: str = ''

run_type: str = 'chat'

token: str = ''

classmethod from_dict(config_dict: dict ) → InferenceConfig

Create InferenceConfig from dictionary (loaded from YAML).

Parameters:

config_dict (dict ) – Configuration dictionary loaded from YAML.

Returns:

InferenceConfig instance.

Return type:

InferenceConfig

validate_all(all_nodes: dict ) → None

Validate all configuration sections.

Parameters:

all_nodes (dict ) – Dictionary of all available nodes in the cluster. Keys are hostnames, values are dragon.native.machine.Node objects.

Raises:

ValueError – If any configuration parameter is invalid.

__init__(model: ModelConfig, hardware: HardwareConfig = <factory>, batching: BatchingConfig = <factory>, guardrails: GuardrailsConfig = <factory>, dynamic_worker: DynamicWorkerConfig = <factory>, flask_secret_key: str = '', run_type: str = 'chat', token: str = '') → None