PromptSnapshotConfig

PromptSnapshotConfig

Configuration types for prompt deployments in the Python SDK.

PromptSnapshotConfig

The model and provider configuration returned as part of a deployment’s prompt snapshot.

from adaline_api.models.prompt_snapshot_config import PromptSnapshotConfig

Fields

provider_name · str · required

The AI provider name (e.g., "openai", "anthropic", "google", "azure", "bedrock").

provider_id · str · required

Adaline’s internal provider identifier.

model · str · required

The model identifier (e.g., "gpt-4o", "claude-sonnet-4-20250514").

settings · dict · required

Normalized model settings. Adaline uses provider-agnostic parameter names that map to each provider’s native parameters.

Settings

The settings dict uses Adaline’s normalized parameter names. When you configure a prompt in Adaline, these are automatically mapped to the correct provider-specific parameter names.

Adaline Key	OpenAI	Anthropic	Google	Description
temperature	temperature	temperature	temperature	Sampling temperature
maxTokens	max_tokens	max_tokens	maxOutputTokens	Maximum output tokens
topP	top_p	top_p	topP	Nucleus sampling threshold
stopSequences	stop	stop_sequences	stopSequences	Stop sequences

---

Examples

Accessing Config from a Deployment

from adaline.main import Adaline

adaline = Adaline()

deployment = await adaline.get_latest_deployment(
    prompt_id="prompt_123",
    deployment_environment_id="environment_123"
)

config = deployment.prompt.config

print(f"Provider: {config.provider_name}")   # e.g. "openai"
print(f"Provider ID: {config.provider_id}")  # e.g. "provider_abc123"
print(f"Model: {config.model}")              # e.g. "gpt-4o"
print(f"Settings: {config.settings}")        # e.g. {"temperature": 0.7, "maxTokens": 1000}

temperature = config.settings.get("temperature")
max_tokens = config.settings.get("maxTokens")

Using Config with a Provider SDK

from adaline.main import Adaline
from openai import OpenAI

adaline = Adaline()
openai = OpenAI()

deployment = await adaline.get_latest_deployment(
    prompt_id="prompt_123",
    deployment_environment_id="environment_123"
)

config = deployment.prompt.config
settings = config.settings

response = openai.chat.completions.create(
    model=config.model,
    messages=messages,
    temperature=settings.get("temperature"),
    max_tokens=settings.get("maxTokens"),
    top_p=settings.get("topP"),
)