Skip to content

test(models): add vLLM and LiteLLM provider contract tests - #22

Open
GulrezQayyum wants to merge 1 commit into
ML4SCI:mainfrom
GulrezQayyum:test/vllm-litellm-contracts
Open

GulrezQayyum wants to merge 1 commit into
ML4SCI:mainfrom
GulrezQayyum:test/vllm-litellm-contracts

Conversation

@GulrezQayyum

@GulrezQayyum GulrezQayyum commented Oct 23, 2026 •

Copy link
Copy Markdown

Summary

Related to #19.

Adds offline contract tests for the existing VLLMModel and LiteLLMModel wrappers.

Coverage

  • default vLLM model name and endpoint;
  • custom vLLM model name and port;
  • default LiteLLM model name and endpoint;
  • custom LiteLLM model name and port.

The tests do not require a running vLLM or LiteLLM server, credentials, or network access.

Documentation

Updates the model integration documentation with:

  • default local endpoints;
  • custom model and port examples;
  • clarification that wrapper construction does not contact a server;
  • distinction between offline contract tests and live provider testing.

Verification

  • uv run pytest -q — 125 passed
  • uv run ruff check src/dlens/agents/_models.py tests/test_models.py
  • git diff --check

Collaboration

The network-request construction test is covered separately in PR #23 to avoid overlapping changes.

This PR focuses on provider configuration and documentation, while #23 verifies that constructing the wrappers does not send HTTP requests.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant