Implementation:Explodinggradients Ragas OCIGenAIWrapper Class

Knowledge Sources	Explodinggradients_Ragas
Domains	LLM, Oracle_Cloud
Last Updated	2026-02-10 00:00 GMT

Overview

Direct integration wrapper for Oracle Cloud Infrastructure (OCI) Generative AI services implementing the Ragas LLM interface.

Description

OCIGenAIWrapper extends BaseRagasLLM to provide text generation using OCI Generative AI services. It supports both sync and async generation, role-aware message conversion, and lazy client initialization. A factory function oci_genai_factory provides convenient instantiation.

Usage

Use this wrapper when running Ragas evaluations with Oracle Cloud's Generative AI models.

Code Reference

Source Location

• Repository: Explodinggradients_Ragas
• File: src/ragas/llms/oci_genai_wrapper.py
• Lines: 35-359

Signature

class OCIGenAIWrapper(BaseRagasLLM):
    def __init__(
        self,
        model_id: str,
        compartment_id: str,
        config: Optional[Dict] = None,
        endpoint_id: Optional[str] = None,
        run_config: Optional[RunConfig] = None,
        cache: Optional[Any] = None,
        default_system_prompt: Optional[str] = None,
        client: Optional[Any] = None,
    ):
        ...
    def generate_text(self, prompt: PromptValue, n: int = 1, temperature: Optional[float] = None, stop: Optional[List[str]] = None, callbacks: Optional[Any] = None) -> LLMResult:
        ...
    async def agenerate_text(self, prompt: PromptValue, n: int = 1, temperature: Optional[float] = None, stop: Optional[List[str]] = None, callbacks: Optional[Any] = None) -> LLMResult:
        ...

def oci_genai_factory(model_id: str, compartment_id: str, ...) -> OCIGenAIWrapper:
    ...

Import

from ragas.llms.oci_genai_wrapper import OCIGenAIWrapper, oci_genai_factory

I/O Contract

Inputs

Name	Type	Required	Description
model_id	str	Yes	OCI model identifier
compartment_id	str	Yes	OCI compartment OCID
config	Optional[Dict]	No	OCI SDK configuration
prompt	PromptValue	Yes	LangChain prompt value to generate from

Outputs

Name	Type	Description
generate_text returns	LLMResult	Generated text with metadata
agenerate_text returns	LLMResult	Async generated text with metadata

Usage Examples

from ragas.llms.oci_genai_wrapper import oci_genai_factory

llm = oci_genai_factory(
    model_id="cohere.command-r-plus",
    compartment_id="ocid1.compartment.oc1...",
)

Related Pages

• Principle:Explodinggradients_Ragas_LLM_Provider_Abstraction