BerriAI/litellm · error · Exception

Unmapped model. Received={}. Expected={}

Error message

Unmapped model. Received={}. Expected={}

What it means

Inside the amazon/titan batch loop, the model string did not equal any of the four hard-coded titan ids (amazon.titan-embed-image-v1, amazon.titan-embed-text-v1, amazon.titan-embed-text-v2:0, amazon.titan-embed-g1-text-02), so no request transformer could be chosen. Note the message template uses {} placeholders — it renders literally instead of interpolating, which is a minor logging bug in the source.

Source

Thrown at litellm/llms/bedrock/embed/embedding.py:440

                    transformed_request: AmazonEmbeddingRequest = (
                        AmazonTitanMultimodalEmbeddingG1Config()._transform_request(
                            input=i, inference_params=inference_params
                        )
                    )
                elif model == "amazon.titan-embed-text-v1":
                    transformed_request = AmazonTitanG1Config()._transform_request(
                        input=i, inference_params=inference_params
                    )
                elif model == "amazon.titan-embed-text-v2:0":
                    transformed_request = AmazonTitanV2Config()._transform_request(
                        input=i, inference_params=inference_params
                    )
                elif model == "amazon.titan-embed-g1-text-02":
                    transformed_request = AmazonTitanG1Config()._transform_request(
                        input=i, inference_params=inference_params
                    )
                else:
                    raise Exception(
                        "Unmapped model. Received={}. Expected={}".format(
                            model,
                            [
                                "amazon.titan-embed-image-v1",
                                "amazon.titan-embed-text-v1",
                                "amazon.titan-embed-text-v2:0",
                                "amazon.titan-embed-g1-text-02",
                            ],
                        )
                    )
                batch_data.append(transformed_request)
        elif provider == "twelvelabs":
            batch_data = []
            for i in input:
                twelvelabs_request = TwelveLabsMarengoEmbeddingConfig()._transform_request(
                    input=i,
                    inference_params=inference_params,
                    async_invoke_route=has_async_invoke,

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Use one of the four exact ids listed in the error: amazon.titan-embed-image-v1, amazon.titan-embed-text-v1, amazon.titan-embed-text-v2:0, amazon.titan-embed-g1-text-02.
2. Strip/normalize the model string (whitespace, case) before calling.
3. Upgrade litellm if AWS shipped a new titan embedding model.

Example fix

# before
resp = litellm.embedding(model="bedrock/amazon.titan-embed-text-v1 ", input=["hi"])

# after
resp = litellm.embedding(model="bedrock/amazon.titan-embed-text-v2:0", input=["hi"])

TITAN_MODELS = {"amazon.titan-embed-image-v1", "amazon.titan-embed-text-v1", "amazon.titan-embed-text-v2:0", "amazon.titan-embed-g1-text-02"}
assert model.strip() in TITAN_MODELS, f"unsupported titan id: {model!r}"

def is_known_titan_id(model: str) -> bool:
    return model.strip() in {
        "amazon.titan-embed-image-v1",
        "amazon.titan-embed-text-v1",
        "amazon.titan-embed-text-v2:0",
        "amazon.titan-embed-g1-text-02",
    }

When it happens

Trigger: Provider inference routed to the amazon family (model contains 'titan'/'amazon') but the exact id differs — e.g. 'amazon.titan-embed-text-v2:0 ' with whitespace, wrong casing, a regional suffix, or a new titan variant not in the list.

Common situations: Copy-pasting model ids with trailing whitespace or unicode quotes; newly launched titan embedding versions on an older litellm; case-mismatched ids.

Related errors

• No embedding data found in response: {response}
• Unable to determine bedrock embedding provider for model: {m
• Bedrock Invoke HTTPX: Unknown provider={provider}, model={mo
• Error processing={raw_response.text}, Received error={e}
• Invalid data URL format: {data_url[:50]}...

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/a4ac2c57563db19e. Report an issue: GitHub.