Update output format of MosaicML endpoint to be more flexible (langch…

…ain-ai#6060)

There will likely be another change or two coming over the next couple
weeks as we stabilize the API, but putting this one in now which just
makes the integration a bit more flexible with the response output
format.

```
(langchain) danielking@MML-1B940F4333E2 langchain % pytest tests/integration_tests/llms/test_mosaicml.py tests/integration_tests/embeddings/test_mosaicml.py 
=================================================================================== test session starts ===================================================================================
platform darwin -- Python 3.10.11, pytest-7.3.1, pluggy-1.0.0
rootdir: /Users/danielking/github/langchain
configfile: pyproject.toml
plugins: asyncio-0.20.3, mock-3.10.0, dotenv-0.5.2, cov-4.0.0, anyio-3.6.2
asyncio: mode=strict
collected 12 items                                                                                                                                                                        

tests/integration_tests/llms/test_mosaicml.py ......                                                                                                                                [ 50%]
tests/integration_tests/embeddings/test_mosaicml.py ......                                                                                                                          [100%]

=================================================================================== slowest 5 durations ===================================================================================
4.76s call     tests/integration_tests/llms/test_mosaicml.py::test_retry_logic
4.74s call     tests/integration_tests/llms/test_mosaicml.py::test_mosaicml_llm_call
4.13s call     tests/integration_tests/llms/test_mosaicml.py::test_instruct_prompt
0.91s call     tests/integration_tests/llms/test_mosaicml.py::test_short_retry_does_not_loop
0.66s call     tests/integration_tests/llms/test_mosaicml.py::test_mosaicml_extra_kwargs
=================================================================================== 12 passed in 19.70s ===================================================================================
```

#### Who can review?

  @hwchase17
  @dev2049

langchain/embeddings/mosaicml.py

@@ -29,7 +29,7 @@ class MosaicMLInstructorEmbeddings(BaseModel, Embeddings):
     """
 
     endpoint_url: str = (
-        "https://models.hosted-on.mosaicml.hosting/instructor-large/v1/predict"
+        "https://models.hosted-on.mosaicml.hosting/instructor-xl/v1/predict"
     )
     """Endpoint URL to use."""
     embed_instruction: str = "Represent the document for retrieval: "
@@ -98,11 +98,38 @@ def _embed(
                     f"Error raised by inference API: {parsed_response['error']}"
                 )
 
-            if "data" not in parsed_response:
-                raise ValueError(
-                    f"Error raised by inference API, no key data: {parsed_response}"
-                )
-            embeddings = parsed_response["data"]
+            # The inference API has changed a couple of times, so we add some handling
+            # to be robust to multiple response formats.
+            if isinstance(parsed_response, dict):
+                if "data" in parsed_response:
+                    output_item = parsed_response["data"]
+                elif "output" in parsed_response:
+                    output_item = parsed_response["output"]
+                else:
+                    raise ValueError(
+                        f"No key data or output in response: {parsed_response}"
+                    )
+
+                if isinstance(output_item, list) and isinstance(output_item[0], list):
+                    embeddings = output_item
+                else:
+                    embeddings = [output_item]
+            elif isinstance(parsed_response, list):
+                first_item = parsed_response[0]
+                if isinstance(first_item, list):
+                    embeddings = parsed_response
+                elif isinstance(first_item, dict):
+                    if "output" in first_item:
+                        embeddings = [item["output"] for item in parsed_response]
+                    else:
+                        raise ValueError(
+                            f"No key data or output in response: {parsed_response}"
+                        )
+                else:
+                    raise ValueError(f"Unexpected response format: {parsed_response}")
+            else:
+                raise ValueError(f"Unexpected response type: {parsed_response}")
+
         except requests.exceptions.JSONDecodeError as e:
             raise ValueError(
                 f"Error raised by inference API: {e}.\nResponse: {response.text}"

langchain/llms/mosaicml.py

@@ -157,18 +157,45 @@ def _call(
                     f"Error raised by inference API: {parsed_response['error']}"
                 )
 
-            if "data" not in parsed_response:
-                raise ValueError(
-                    f"Error raised by inference API, no key data: {parsed_response}"
-                )
-            generated_text = parsed_response["data"]
+            # The inference API has changed a couple of times, so we add some handling
+            # to be robust to multiple response formats.
+            if isinstance(parsed_response, dict):
+                if "data" in parsed_response:
+                    output_item = parsed_response["data"]
+                elif "output" in parsed_response:
+                    output_item = parsed_response["output"]
+                else:
+                    raise ValueError(
+                        f"No key data or output in response: {parsed_response}"
+                    )
+
+                if isinstance(output_item, list):
+                    text = output_item[0]
+                else:
+                    text = output_item
+            elif isinstance(parsed_response, list):
+                first_item = parsed_response[0]
+                if isinstance(first_item, str):
+                    text = first_item
+                elif isinstance(first_item, dict):
+                    if "output" in parsed_response:
+                        text = first_item["output"]
+                    else:
+                        raise ValueError(
+                            f"No key data or output in response: {parsed_response}"
+                        )
+                else:
+                    raise ValueError(f"Unexpected response format: {parsed_response}")
+            else:
+                raise ValueError(f"Unexpected response type: {parsed_response}")
+
+            text = text[len(prompt) :]
+
         except requests.exceptions.JSONDecodeError as e:
             raise ValueError(
                 f"Error raised by inference API: {e}.\nResponse: {response.text}"
             )
 
-        text = generated_text[0][len(prompt) :]
-
         # TODO: replace when MosaicML supports custom stop tokens natively
         if stop is not None:
             text = enforce_stop_tokens(text, stop)