Vertex AI
No written verdict yet. Here is what commenters said.
Why people recommend it
“the main provider i know that just doesn't go down is vertex”
“I use Gemini via the VertexAi API, much better”
Show 3 more
“just use those for direct Gemini/Vertex API access”
“I pay by the token via VertexAi, better than those budget plans that rate limit you”
“the (theoretical) production platform for Gemini is Vertex AI, not AI Studio”
Caveats and dissent
“Add that to AI studio and Vertex's famously terrible caching”
“default long-term memory uses vertext AI (gemini) in ADK.”
Show 2 more
“GCP/Vertex/whatever they're offering is a billing nightmare”
“Google Vertex is even worse.”
Often compared with
Amazon BedrockWho recommends it in their own words
Threads where it came up
- DeepSeek-v3.2: Pushing the frontier of open large language models [pdf] 3 mentions
- Ask HN: Who wants to be hired? (September 2026) 2 mentions
- The obscure trust looming over Anthropic's IPO 1 mention
- Just say no to OpenAI and Anthropic 1 mention
- Gemini 4 Argon 1 mention
- OpenAI Agents API 1 mention
Also written as Google Vertex AI, Google Vertex. Data: vertex-ai.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 36
“we use a copy hosted on a GCP Vertex AI instance”
“being offered as providers with zero data retention”
“Add that to AI studio and Vertex's famously terrible caching”
“default long-term memory uses vertext AI (gemini) in ADK.”
“proxying Google Vertex AI image generation”
“When I use Vertex or the Bedrock, my AI API requests from my instances doesn't leave their networks.”
“Vertex AI”
“GCP/Vertex/whatever they're offering is a billing nightmare”
“Google Vertex is even worse.”
“through Vertex [1]”
“to access all VertexAI models we need to support a bunch”
“Vertex”
“Vertex / Bedrock have zero prompt logging”
“Google Vertex (or Google AI Studio) might be a potential alternative, but the UX is a lot worse”
“even though Vertex is mostly used for inference”
“LLMs (OpenAI, Claude, Vertex AI, Llama)”
“for clients we use paid Vertex AI accounts”
“the main provider i know that just doesn't go down is vertex”
“for which I switch to API billing on Vertex”
“Stack: Langgraph, Langfuse, Go, Python, TS, GraphQL, gRPC/Connect-RPC, PostgreSQL, GCP/Vertex AI, AWS Bedrock.”
“I use Gemini via the VertexAi API, much better”
“just use those for direct Gemini/Vertex API access”
“which uses vertex ai and all models therein”
“I pay by the token via VertexAi, better than those budget plans that rate limit you”
“AI: Vertex AI”
“Vertex AI (Gemini 1.5/Pro)”
“I don't see anything you do that they don't already do for me.”
“or Google models through Vertex API”
“Also built a Google Vertex AI integration demo”
“Gemini (both UI & Vertex for initial Annual Town Report summaries & highlights)”
“we put Google's Vertex AI product discovery algorithms on Shopify and Salesforce ecommerce storefronts”
“the (theoretical) production platform for Gemini is Vertex AI, not AI Studio”
“I don't see any large base models there.”
GCloud and Hyperbolic have been my go-to as well
“If you’re already using GCP, Vertex AI is pretty good.”
“load your stuff into Vertex or whatever”