Auto Detection¶

LLM-Rosetta can automatically detect which provider format a request uses.

Detecting Provider¶

from llm_rosetta import detect_provider

provider = detect_provider(request_dict)
# Returns: "openai_chat", "openai_responses", "anthropic", "google", or None

Getting a Converter¶

from llm_rosetta import get_converter_for_provider

converter = get_converter_for_provider("anthropic")

Convenience Conversion¶

from llm_rosetta import convert

# Auto-detect source, convert to target
result = convert(
    source_body=openai_request,
    target_provider="anthropic",
    source_provider=None,  # auto-detect
)

Force Conversion for Same-Provider Requests¶

By default, convert() returns the body as-is when the source and target providers are the same. Use force_conversion=True to run the full conversion pipeline even in this case — useful for parameter normalization:

# Normalize OpenAI Chat parameters (e.g. max_tokens → max_completion_tokens)
result = convert(
    source_body=openai_request,
    target_provider="openai_chat",
    force_conversion=True,
)