मुख्य सामग्री पर जाएं

Installation

pip install rail-score-sdk
Provider integrations के लिए optional extras के साथ install करें:
pip install "rail-score-sdk[openai]"       # OpenAI wrapper
pip install "rail-score-sdk[anthropic]"    # Anthropic wrapper
pip install "rail-score-sdk[google]"       # Gemini wrapper
pip install "rail-score-sdk[litellm]"      # LiteLLM wrapper
pip install "rail-score-sdk[integrations]" # सभी LLM providers
pip install "rail-score-sdk[agents]"       # CrewAI, LangGraph, AutoGen
pip install "rail-score-sdk[telemetry]"    # OpenTelemetry support
pip install "rail-score-sdk[dev]"          # Development tools

Sync client

from rail_score_sdk import RailScoreClient

client = RailScoreClient(api_key="YOUR_RAIL_API_KEY")

result = client.eval(content="Your AI-generated text here", mode="basic")
print(f"RAIL Score: {result.rail_score.score}/10")
RailScoreClient typed dataclass objects return करता है। Scores को access करें जैसे result.rail_score.score, result.dimension_scores["fairness"].score, वगैरह।

Async client

import asyncio
from rail_score_sdk import AsyncRAILClient

async def main():
    client = AsyncRAILClient(api_key="YOUR_RAIL_API_KEY")
    result = await client.eval(content="Your text here", mode="basic")
    print(result["rail_score"]["score"])  # Raw dicts return करता है

asyncio.run(main())
AsyncRAILClient raw dictionaries return करता है, dataclasses नहीं।

Key classes

ClassPurpose
RailScoreClientSync client - सभी core methods
AsyncRAILClientAsync client - सभी core methods
RAILSessionConversation में quality track करें
PolicyScore enforcement के लिए declarative rules
RuleIndividual policy rule
RAILMiddlewareकिसी भी async LLM function को wrap करें

Error handling

from rail_score_sdk import (
    RailScoreClient,
    AuthenticationError,
    InsufficientCreditsError,
    RateLimitError,
    ContentTooHarmfulError,
)

client = RailScoreClient(api_key="YOUR_RAIL_API_KEY")

try:
    result = client.eval(content="...", mode="deep")
except AuthenticationError:
    print("API key check करें")
except InsufficientCreditsError as e:
    print(f"{e.required} credits चाहिए, {e.balance} हैं")
except RateLimitError:
    print("Requests slow करें")
except ContentTooHarmfulError:
    print("Content safety layer पर block हो गया")

आगे क्या देखें

Evaluation

Sync और async eval examples।

Safe Regeneration

Threshold से नीचे वाले content को auto-fix करें।

Sessions & Policy

Conversations में quality track करें।

Integrations

OpenAI, Gemini, Anthropic के लिए provider wrappers।