A Ruby gem for seamlessly and uniformly interacting with large language and vision model (LLM) API's served by numerous services, including those of OpenAI, Anthropic, Google and others.
Groq will honor a stop sequence when streaming but it seems to continue to generate in the backend and, should it hit max_token, it will return length as the finish reason rather than stop.
Groq will honor a stop sequence when streaming but it seems to continue to generate in the backend and, should it hit
max_token, it will returnlengthas the finish reason rather thanstop.