Available Models
Chat Models
| Model | Model ID | Context | Best For |
|---|---|---|---|
| GPT-4o | gpt-4o | 128K tokens | Fast general-purpose coding, broad knowledge |
| GPT-4o Mini | gpt-4o-mini | 128K tokens | Cost-effective lightweight tasks |
| GPT-4 Turbo | gpt-4-turbo | 128K tokens | Complex coding with vision support |
Reasoning Models (o-series)
| Model | Model ID | Context | Best For |
|---|---|---|---|
| o3 | o3 | 200K tokens | Hardest reasoning, math, complex code architecture |
| o3-mini | o3-mini | 200K tokens | Fast reasoning at lower cost |
| o4-mini | o4-mini | 200K tokens | Newest reasoning model, balanced speed/capability |
| o1 | o1 | 200K tokens | Deep multi-step reasoning |
Note
The o-series models use extended thinking internally. Responses may take longer but produce higher quality results for complex problems. They are particularly strong at multi-step code reasoning.
Setup
Step 1: Get an API Key
- Go to platform.openai.com and sign in or create an account.
- Navigate to API Keys under your account settings.
- Click "Create new secret key" and copy it (it starts with sk-).
- Make sure your account has a payment method and sufficient credits.
Step 2: Add the Key to Creor
Option A: Settings UI (recommended)
- Open Creor and go to Settings.
- Navigate to the Providers section.
- Find OpenAI and enter your API key.
- The key is stored securely in your OS keychain.
Option B: Environment Variable
export OPENAI_API_KEY="sk-your-key-here"
Configuration
Set an OpenAI model as your default in creor.json:
1
2
3
{
"model": "openai/gpt-4o"
}
Use a reasoning model for the plan agent and a fast model for building:
1
2
3
4
5
6
7
8
9
10
{
"agent": {
"build": {
"model": "openai/gpt-4o"
},
"plan": {
"model": "openai/o3"
}
}
}
Advanced Options
Custom Base URL
If you use an OpenAI-compatible proxy, configure a custom base URL:
1
2
3
4
5
6
7
8
9
{
"provider": {
"openai": {
"options": {
"baseURL": "https://your-openai-proxy.example.com/v1"
}
}
}
}
Organization ID
If your OpenAI account belongs to an organization, set the org ID:
export OPENAI_ORG_ID="org-your-org-id"
Model Filtering
1
2
3
4
5
6
7
{
"provider": {
"openai": {
"whitelist": ["gpt-4o", "o3-mini"]
}
}
}
Best For
- Fast, general-purpose coding with broad domain knowledge.
- Quick iteration loops where response speed matters.
- Multi-step reasoning tasks (o-series models).
- Projects that need GPT-specific features like vision input.
- Teams already using OpenAI for other products.