Google AI Studio (Gemini)

Configuring Google AI Studio in Aimogen Pro for Gemini text, vision, embeddings and Imagen image generation, plus the free-tier limits to expect.

Google AI Studio gives Aimogen Pro access to the Gemini family for text, vision, streaming, tool calling and embeddings, plus Imagen and Gemini image models for image generation.

Get a key#

  1. Go to aistudio.google.com/app/apikey.
  2. Create an API key, optionally attached to a Google Cloud project.
  3. For production volume, attach a billing-enabled project. Keys without one are limited to the free tier.

Configure#

Aimogen Pro › Settings › API Keys › Google AI Studio AI API Keys

One key per line. Select Test API key, then save. Gemini models then appear in every model selector.

Models#

Aimogen Pro 2.8.8 lists 20 Google text models.

ModelNotes
gemini-3.7-flash, gemini-3.6-flashAdded in 2.8.9
gemini-3.5-flashAdded in 2.8.1
gemini-3.1-flash-lite-preview, gemini-3-flash-preview, gemini-3.1-pro-previewGemini 3 preview line
gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-liteGemini 2.5
gemini-2.0-flash, gemini-2.0-flash-lite, gemini-2.0-pro-exp-02-05, gemini-2.0-flash-exp, gemini-2.0-flash-thinking-exp-01-21Gemini 2.0
gemma-3-27b-itOpen-weight Gemma
chat-bison-001, text-bison-001Legacy PaLM, retained for backward compatibility

The plugin maintains separate capability lists for vision, streaming, chat and function calling. Current Gemini models are in all of them; the legacy Bison models are not.

Embeddings#

Google is one of the embedding providers Aimogen Pro supports:

Model
gemini-embedding-2-preview
gemini-embedding-001
text-embedding-005

Configure under Settings › Embeddings. See Embedding models.

Image generation#

Google image models are available for image generation:

Model
imagen-4.0-generate-001 (default)
gemini-3.1-flash-image-preview
gemini-3-pro-image-preview
gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite

Set them under Settings › AI Images, or per rule.

Free tier#

Google offers a genuinely usable free tier, which makes Gemini a good first provider for evaluation. It comes with tight per-minute and per-day request limits.

If bulk runs fail with RESOURCE_EXHAUSTED or HTTP 429, you are hitting them. Options:

  • Raise Delay Between API & Scraping Requests (ms)
  • Attach a billing-enabled project
  • Add more keys to the textarea and let rotation spread the load

Safety filters#

Gemini applies safety filters and can return an empty response with a SAFETY finish reason rather than an error. Aimogen Pro classifies this as a safety failure, which Reliability Mode deliberately does not cool down or retry on another model — retrying a blocked prompt on a different model usually blocks again.

If legitimate content is being blocked, rewrite the prompt. Marketing copy about weapons, health claims, alcohol and similar topics trips these filters most often.

Common problems#

API_KEY_INVALID The key is wrong, or the Generative Language API is not enabled on the project.

RESOURCE_EXHAUSTED / 429 Free-tier quota. See above.

Empty response, no error A safety block. Check System & Logs › Activity Logs for the finish reason.

A model in the dropdown returns 404 Preview models are region-limited and are sometimes withdrawn. Choose a stable model.

Endpoint used#

https://generativelanguage.googleapis.com/v1beta/...

Still stuck? Open a support ticket and include the diagnostics from Aimogen Pro › System & Logs › System Info.