Home / File/ test_llms.py — langchain Source File

test_llms.py — langchain Source File

Architecture documentation for test_llms.py, a python file in the langchain codebase. 4 imports, 0 dependents.

Entity Profile

Dependency Diagram

graph LR
  a02e3dbf_01a6_cd9e_055a_f10c341db951["test_llms.py"]
  8e2034b7_ceb8_963f_29fc_2ea6b50ef9b3["typing"]
  a02e3dbf_01a6_cd9e_055a_f10c341db951 --> 8e2034b7_ceb8_963f_29fc_2ea6b50ef9b3
  6e58aaea_f08e_c099_3cc7_f9567bfb1ae7["pydantic"]
  a02e3dbf_01a6_cd9e_055a_f10c341db951 --> 6e58aaea_f08e_c099_3cc7_f9567bfb1ae7
  120e2591_3e15_b895_72b6_cb26195e40a6["pytest"]
  a02e3dbf_01a6_cd9e_055a_f10c341db951 --> 120e2591_3e15_b895_72b6_cb26195e40a6
  9d19aeb9_35aa_ec9e_adca_9ac1aea84236["langchain_fireworks"]
  a02e3dbf_01a6_cd9e_055a_f10c341db951 --> 9d19aeb9_35aa_ec9e_adca_9ac1aea84236
  style a02e3dbf_01a6_cd9e_055a_f10c341db951 fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

"""Test Fireworks LLM."""

from typing import cast

from pydantic import SecretStr
from pytest import CaptureFixture, MonkeyPatch

from langchain_fireworks import Fireworks


def test_fireworks_api_key_is_secret_string() -> None:
    """Test that the API key is stored as a SecretStr."""
    llm = Fireworks(  # type: ignore[call-arg]
        fireworks_api_key="secret-api-key",
        model="accounts/fireworks/models/mixtral-8x7b-instruct",
        temperature=0.2,
        max_tokens=250,
    )
    assert isinstance(llm.fireworks_api_key, SecretStr)

    # Test api_key alias
    llm = Fireworks(
        api_key="secret-api-key",  # type: ignore[arg-type]
        model="accounts/fireworks/models/mixtral-8x7b-instruct",
        temperature=0.2,
        max_tokens=250,
    )
    assert isinstance(llm.fireworks_api_key, SecretStr)


def test_fireworks_api_key_masked_when_passed_from_env(
    monkeypatch: MonkeyPatch, capsys: CaptureFixture
) -> None:
    """Test that the API key is masked when passed from an environment variable."""
    monkeypatch.setenv("FIREWORKS_API_KEY", "secret-api-key")
    llm = Fireworks(
        model="accounts/fireworks/models/mixtral-8x7b-instruct",
        temperature=0.2,
        max_tokens=250,
    )
    print(llm.fireworks_api_key, end="")  # noqa: T201
    captured = capsys.readouterr()

    assert captured.out == "**********"


def test_fireworks_api_key_masked_when_passed_via_constructor(
    capsys: CaptureFixture,
) -> None:
    """Test that the API key is masked when passed via the constructor."""
    llm = Fireworks(  # type: ignore[call-arg]
        fireworks_api_key="secret-api-key",
        model="accounts/fireworks/models/mixtral-8x7b-instruct",
        temperature=0.2,
        max_tokens=250,
    )
    print(llm.fireworks_api_key, end="")  # noqa: T201
    captured = capsys.readouterr()

    assert captured.out == "**********"


def test_fireworks_uses_actual_secret_value_from_secretstr() -> None:
    """Test that the actual secret value is correctly retrieved."""
    llm = Fireworks(  # type: ignore[call-arg]
        fireworks_api_key="secret-api-key",
        model="accounts/fireworks/models/mixtral-8x7b-instruct",
        temperature=0.2,
        max_tokens=250,
    )
    assert cast(SecretStr, llm.fireworks_api_key).get_secret_value() == "secret-api-key"


def test_fireworks_model_params() -> None:
    # Test standard tracing params
    llm = Fireworks(model="foo", api_key="secret-api-key")  # type: ignore[arg-type]

    ls_params = llm._get_ls_params()
    assert ls_params == {
        "ls_provider": "fireworks",
        "ls_model_type": "llm",
        "ls_model_name": "foo",
    }

    llm = Fireworks(
        model="foo",
        api_key="secret-api-key",  # type: ignore[arg-type]
        max_tokens=10,
        temperature=0.1,
    )

    ls_params = llm._get_ls_params()
    assert ls_params == {
        "ls_provider": "fireworks",
        "ls_model_type": "llm",
        "ls_model_name": "foo",
        "ls_max_tokens": 10,
        "ls_temperature": 0.1,
    }

Subdomains

Dependencies

  • langchain_fireworks
  • pydantic
  • pytest
  • typing

Frequently Asked Questions

What does test_llms.py do?
test_llms.py is a source file in the langchain codebase, written in python. It belongs to the CoreAbstractions domain, RunnableInterface subdomain.
What functions are defined in test_llms.py?
test_llms.py defines 5 function(s): test_fireworks_api_key_is_secret_string, test_fireworks_api_key_masked_when_passed_from_env, test_fireworks_api_key_masked_when_passed_via_constructor, test_fireworks_model_params, test_fireworks_uses_actual_secret_value_from_secretstr.
What does test_llms.py depend on?
test_llms.py imports 4 module(s): langchain_fireworks, pydantic, pytest, typing.
Where is test_llms.py in the architecture?
test_llms.py is located at libs/partners/fireworks/tests/unit_tests/test_llms.py (domain: CoreAbstractions, subdomain: RunnableInterface, directory: libs/partners/fireworks/tests/unit_tests).

Analyze Your Own Codebase

Get architecture documentation, dependency graphs, and domain analysis for your codebase in minutes.

Try Supermodel Free