Re:Google Gemini has the worst LLM API
by lemming on Saturday May 03, 2025 @08:23PM
(#43883464)
Additionally, there's no OpenAPI spec, so you have to generate one from their protobuf specs if you want to use that to generate a client model. Their protobuf specs live in a repo at https://github.com/googleapis/googleapis/tree/master/google/.... Now you might think that v1 would be the latest there, but you would be wrong - everyone uses v1beta (not v1, not v1alpha, not v1beta3) for reasons that are completely unclear. Additionally, this repo is frequently not up to date with the actual API (it took them ages to get the new thinking config added, for example, and their usage fields were out of date for the longest time). It's really frustrating.
[ Parent
| on HN ]
Eh, you know. "Move fast and break things."
Re:Google Gemini has the worst LLM API
by caturopath on Saturday May 03, 2025 @10:09PM
(#43883933)
I'm not sure "move fast" describes the situation.
Hmm, the proliferation of branches, including some which seem perhaps more recent than "v1beta" made me imagine this could apply.
Re:Google Gemini has the worst LLM API
by chrisheecho on Sunday May 04, 2025 @02:48AM
(#43885002)
lemming, this is super helpful, thank you. We provide the genai SDK (https://github.com/googleapis/python-genai) to reduce the learning curve in 4 languages (GA: Python, Go Preview: Node.JS, Java). The SDK works for all Gemini APIs provided by Google AI Studio (https://ai.google.dev/) and Vertex AI.
Re:Google Gemini has the worst LLM API
by egamirorrim on Sunday May 04, 2025 @03:27AM
(#43885167)
The way dependency resolution works in Java with the special, Google only, giant dynamic BOM resolver is hell on earth.
We have to write code that round robins every region on retries to get past how overloaded/poorly managed vertex is (we're not hitting our quotas) and yes that's even with retry settings on the SDK.
Read timeouts aren't configurable on the Vertex SDK.