building an ai product or running agents? you can now call google's ai models just like you'd call openai.
right now you can hit:
>gemini-3.5-flash
>gemini-3.1-flash-image
>gemini-3-pro-image
>veo-3.1-generate-001
>veo-3.1-fast-generate-001
i built velros llm gateway.
it gives you an openai-compatible base url for google ai models on vertex.
just drop it into whatever you're already running:
>openclaw
>hermes agent
>opencode
>your own custom agents
base url: https://t.co/La1SLQgpzY
sign in with google sso and you get an api key on the spot. no forms, no waiting around.
new beta users get $50 in free credits to mess around with.
reply and tell me what you're building if you want me to bump your use case up the list.
and if you're building something serious, dm me. i'm handing selected builders higher limits plus up to $100 extra in beta credits.