News for hackers. Stuff that matters.

A comment on Google Gemini has the worst LLM API. Parent comment

Re:Google Gemini has the worst LLM API by minimaxir on Saturday May 03, 2025 @08:20PM (#43883445)
From the linked docs:

> If you want to disable thinking, you can set the reasoning effort to "none".

For other APIs, you can set the thinking tokens to 0 and that also works.

[ Parent | on HN ]
Re:Google Gemini has the worst LLM API by mgraczyk on Saturday May 03, 2025 @08:25PM (#43883478)
Wow thanks I did not know
[ Parent | on HN ]
Re:Google Gemini has the worst LLM API by logankilpatrick on Sunday May 11, 2025 @03:14PM (#43956241)
We added it to the docs. The downside of the OAI compat endpoint is we have to design the API twice, once for our API, then once through the OAI compat layer which makes it slower sometimes to have certain features, especially if we diverge at all.
[ Parent | on HN ]
Re:Google Gemini has the worst LLM API by mgraczyk on Sunday May 11, 2025 @03:17PM (#43956271)
Thanks, yes makes sense.

BTW, I have noticed that when tested outside GCP, the OpenAI compat endpoint has significantly lower latency for most requests (vs using the genai library). VertexAI is better than both.

Any idea why or if that will change?

[ Parent | on HN ]