Default models with original names

[kachar]

We're trying to use the docker container to run our e2e tests against a mock endpoints.

We're facing an issue with the mock models as they're not matching the expected original names of models for each provider.

For example with OpenAI we have gpt-4-mock instead of gpt-4o, etc

It would be great if we can adjust/provide/set the models to be used in the '/v1/models endpoints.

Ideas are welcome

[zerob13]

Great idea! I’m planning to provide config file to map the provided models, so you can freely change the names of the several provided models.