Skip to content
ShopAssist Docs
AI

Google Gemini

Use a paid-tier Gemini API key with ShopAssist; free-tier keys are rejected because Google may use their data.

Create a key

As of October 2026, in Google AI Studio:

  1. Open aistudio.google.com → Get API key and create a key in a Google Cloud project.
  2. In the project, enable billing (a paid tier). Free-tier keys are rejected by ShopAssist.
  3. Copy the key (AIza…).

Account tier

Paid tier only. On Google's free tier, prompts and responses may be used to improve Google's products, which is not acceptable for shopper conversations. You confirm the paid tier when saving the key, and the connection test rejects keys whose errors mention the free tier (gemini_free_tier).

Recommended models

Models you can use with your own key
ModelTierContext windowImagesProcessing
Gemini 3.1 Flash-Lite gemini-3.1-flash-liteStandard1,048,576 tokensYesGoogle global
Gemini 3.5 Flash-Lite gemini-3.5-flash-liteStandard1,048,576 tokensYesGoogle global
Gemini 3.8 Flash gemini-3.8-flashPremium1,048,576 tokensYesGoogle global

Fields

FieldNotes
API keyAIza…
Paid tier confirmationRequired
ModelA model from the list or any current gemini- model

Data location

Data location

Only paid-tier keys are accepted. Google doesn't guarantee EU processing on the Gemini API; use Vertex AI for EU residency.

Connection test

Checks run by Test connection
CheckRequiredWhat it testsOn failure
authYesThe key is accepted (lists models or sends a tiny request)auth_failed, insufficient_quota, gemini_free_tier, endpoint_blocked, provider_unavailable, model_not_found
toolsYesThe model calls a tool with valid arguments and continues after the resulttools_unsupported: the model can't run the shopping tools
ttftNo (degraded)The first answer event arrives within 8 secondsttft_slow: works, but shoppers wait
parallel_toolsNoTwo tools are called in one stepparallel_tools_unsupported: slightly slower answers
visionNoThe model names the colour of a test imagevision_unsupported: visual search is off
contextSelf-hosted onlyA fact at the start of an 8,000-token prompt is recalledcontext_too_small: the endpoint truncates prompts

Cost

What your own AI key costs

Estimate based on list prices on 2026-09-28; your provider bills you directly.

Estimated provider cost per 1,000 conversations
ProviderModelUSD per 1,000 conversations≈ EUR per 1,000 conversations
Azure OpenAIGPT-6 Luna$3.61€3.10
MistralMistral Small 4$4.71€4.05
OpenAIGPT-5.6 Luna$7.62€6.55
Google GeminiGemini 3.1 Flash-Lite$8.85€7.61
OpenAIGPT-5 mini$9.85€8.47
Google GeminiGemini 3.5 Flash-Lite$12.02€10.34
MistralMistral Large 3$14.70€12.64
OpenAIGPT-5.4 mini$26.55€22.83
AnthropicClaude Haiku 4.5$36.10€31.05
AnthropicClaude Sonnet 5$72.20€62.09

Troubleshooting

  • gemini_free_tier: enable billing on the key's Google Cloud project and test again.
  • 401 / auth_failed: the key is restricted to other APIs or was deleted; check API restrictions in the Cloud console.
  • 429 / rate_limited or insufficient_quota: raise the project's quota.
  • model_not_found: preview models change names; choose a stable model.

Last updated on

On this page