This PR migrates the codebase from the deprecated Azure inference endpoint to the new GitHub Models API endpoint, as announced in the GitHub Models deprecation notice.

Changes Made

🔗 URL Migration

• Replace all instances of https://models.inference.ai.azure.com with https://models.github.ai/inference
• Updated in:
  • src/backend/fastapi_app/openai_clients.py (2 instances)
  • .env.sample

🏷️ Model Name Updates

• Add required "openai/" prefix to all GitHub Models model names:
  • Chat models: gpt-4o → openai/gpt-4o
  • Embedding models: text-embedding-3-large → openai/text-embedding-3-large
  • Updated defaults in both openai_clients.py and dependencies.py
  • Updated sample configuration in .env.sample

🧪 Test Coverage

• Added comprehensive tests to validate GitHub Models configuration
• Tests verify both default values and custom environment variable overrides
• All existing tests continue to pass

Example Usage

After this change, the correct configuration for GitHub Models is:

if API_HOST == "github":
 client = openai.OpenAI(
 base_url="https://models.github.ai/inference",
 api_key=os.environ["GITHUB_TOKEN"],
 )
 MODEL_NAME = os.getenv("GITHUB_MODEL", "openai/gpt-4o")

Validation

• ✅ All existing functionality preserved
• ✅ New tests validate correct URL and model name usage
• ✅ Code quality checks pass
• ✅ Zero remaining references to deprecated endpoint

Fixes #235.

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.