Available Models
| Model | Model ID | Context | Best For |
|---|---|---|---|
| Gemini 2.5 Pro | gemini-2.5-pro-preview-05-06 | 1M tokens | Complex reasoning, long context, multimodal |
| Gemini 2.5 Flash | gemini-2.5-flash-preview-05-20 | 1M tokens | Fast responses, cost-effective, large context |
| Gemini 3 Flash Preview | gemini-3-flash-preview | 1M tokens | Cutting-edge speed and capability |
| Gemini 2.0 Flash | gemini-2.0-flash | 1M tokens | Stable, production-ready |
Tip
Gemini models support up to 1 million tokens of context, making them excellent for working with very large codebases where you need to include many files in a single conversation.
Google AI Studio
Google AI Studio is the easiest way to get started with Gemini models. It provides free API keys with generous rate limits.
Step 1: Get an API Key
- Go to aistudio.google.com and sign in with your Google account.
- Click "Get API key" in the left sidebar.
- Create a new API key or use an existing one.
- Copy the key.
Step 2: Add the Key to Creor
Option A: Settings UI (recommended)
- Open Creor and go to Settings.
- Navigate to the Providers section.
- Find Google AI Studio and enter your API key.
Option B: Environment Variable
export GOOGLE_GENERATIVE_AI_API_KEY="your-key-here"
Configuration
1
2
3
{
"model": "google/gemini-2.5-pro-preview-05-06"
}
Google Vertex AI
Google Vertex AI is the enterprise option for running Gemini models. It uses your existing GCP project and IAM authentication, making it ideal for organizations with compliance requirements.
Prerequisites
- A Google Cloud Platform project with billing enabled.
- The Vertex AI API enabled in your project.
- Application Default Credentials (ADC) configured on your machine.
Step 1: Set Up Authentication
Configure Application Default Credentials using the gcloud CLI:
gcloud auth application-default login
Step 2: Set Environment Variables
1
2
3
4
5
# Required: your GCP project ID
export GOOGLE_CLOUD_PROJECT="your-project-id"
# Optional: region (defaults to us-east5)
export GOOGLE_CLOUD_LOCATION="us-east5"
Configuration
1
2
3
{
"model": "google-vertex/gemini-2.5-pro-preview-05-06"
}
Note
Note the different provider prefix:
google/ for AI Studio and google-vertex/ for Vertex AI. The model IDs are the same, but the routing and authentication differ.Vertex Anthropic (Claude on GCP)
Google Vertex AI also hosts Claude models from Anthropic. This lets you use Claude while keeping all traffic within your GCP infrastructure.
Setup
1
2
3
4
5
# Same GCP credentials as Vertex AI
export GOOGLE_CLOUD_PROJECT="your-project-id"
# Claude models use "global" as the default location
export GOOGLE_CLOUD_LOCATION="global"
Configuration
1
2
3
{
"model": "google-vertex-anthropic/claude-sonnet-4-20250514"
}
Configuration Examples
AI Studio with Agent Overrides
1
2
3
4
5
6
7
8
9
10
11
12
{
"model": "google/gemini-2.5-pro-preview-05-06",
"small_model": "google/gemini-2.5-flash-preview-05-20",
"agent": {
"build": {
"model": "google/gemini-2.5-pro-preview-05-06"
},
"explore": {
"model": "google/gemini-2.5-flash-preview-05-20"
}
}
}
Vertex AI Enterprise Setup
1
2
3
4
5
6
7
8
9
10
11
12
{
"model": "google-vertex/gemini-2.5-pro-preview-05-06",
"enabled_providers": ["google-vertex", "google-vertex-anthropic"],
"provider": {
"google-vertex": {
"options": {
"project": "my-gcp-project",
"location": "us-east5"
}
}
}
}
Best For
- Working with very large codebases (up to 1M tokens of context).
- Multimodal tasks involving images, diagrams, or screenshots.
- Cost-effective coding with Gemini Flash models.
- Enterprise deployments on Google Cloud (Vertex AI).
- Teams that need Claude models through GCP infrastructure (Vertex Anthropic).