Configuration
Configure SheetMagic settings, models, and preferences
Basic Configuration
SheetMagic settings can be accessed through the sidebar or settings panel in Google Sheets.
Accessing Settings
- Go to Extensions → SheetMagic → Settings
- Or click the gear icon in the SheetMagic sidebar
Provider Configuration
Selecting a Provider
Choose your preferred AI provider from the dropdown:
| Provider | Best For |
|---|---|
| OpenAI | Full-featured: text, images, vision, speech, video |
| Claude | Long documents, complex reasoning |
| Gemini | Large context (1M tokens), free tier |
| OpenRouter | Access to 100+ models |
| Straico | All-in-one with images, video, audio |
API Key Setup
Provider: OpenAI
API Key: sk-proj-xxxxxxxxxxxxxxxxGet your key at platform.openai.com/api-keys
Provider: Claude
API Key: sk-ant-xxxxxxxxxxxxxxxxGet your key at console.anthropic.com/settings/keys
Provider: Gemini
API Key: AIzaxxxxxxxxxxxxxxxxGet your key at aistudio.google.com/app/apikey
Provider: OpenRouter
API Key: sk-or-xxxxxxxxxxxxxxxxGet your key at openrouter.ai/keys
Provider: Straico
API Key: xxxxxxxxxxxxxxxxGet your key at platform.straico.com/user-settings
Model Selection
Text Generation Models
Select your preferred model for =AITEXT(), =AILIST(), and =AILISTH():
| Provider | Recommended Models |
|---|---|
| OpenAI | gpt-5, gpt-4o, gpt-4o-mini (free tier) |
| Claude | claude-opus-4-5, claude-sonnet-4-5, claude-haiku-4-5 |
| Gemini | gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite (free) |
| OpenRouter | openai/gpt-4o, anthropic/claude-opus-4.5 |
| Straico | openai/gpt-5, anthropic/claude-sonnet-4.5 |
Image Generation Models
Select your preferred model for =AIIMAGE():
| Provider | Available Models |
|---|---|
| OpenAI | dall-e-3, dall-e-2, gpt-image-1 |
| Gemini | imagen-4-ultra, imagen-4, imagen-4-fast |
| Straico | dall-e-3, flux-1.1, ideogram-v3 |
Video Generation Models
Select your preferred model for =AIVIDEO():
| Provider | Available Models |
|---|---|
| OpenAI | sora-2-pro, sora-2 |
| Gemini | veo-3, veo-3-fast |
| Straico | veo-2, runway-gen-4-turbo |
Advanced Settings
Temperature
Controls the creativity/randomness of AI responses.
| Value | Behavior |
|---|---|
| 0 | Deterministic, consistent responses |
| 0.5 | Balanced |
| 1 | Default, moderate creativity |
| 2 | Highly creative, varied responses |
Temperature is not supported for reasoning models (o1, o3, o4). These models use reasoning effort levels instead.
Max Tokens
Controls the maximum length of AI responses. Default is 500 tokens.
- Short responses: 100-200 tokens
- Standard responses: 500 tokens (default)
- Long responses: 1000-2000 tokens
- Very long responses: 4000+ tokens
Reasoning Effort (Reasoning Models Only)
For models like o1, o3, and o4:
| Level | Description |
|---|---|
| Low | Quick responses, less reasoning |
| Medium | Balanced reasoning (default) |
| High | Deep reasoning, more thorough |
Function-Specific Settings
AITEXT Settings
Model: gpt-4o
Temperature: 1.0
Max Tokens: 500AIIMAGE Settings
Model: dall-e-3
Size: 1024x1024
Quality: standardAISPEECH Settings
Model: tts-1-hd
Voice: alloy (default)
Speed: 1.0GPTV Settings
Model: gpt-4o
Max Tokens: 500
Detail: autoBest Practices
For Cost Optimization
| Task | Recommended Model |
|---|---|
| Simple queries | gpt-4o-mini, gemini-2.5-flash-lite |
| Standard content | gpt-4o, claude-sonnet-4.5 |
| Complex analysis | gpt-5, claude-opus-4.5 |
| High volume | Haiku models, Flash-Lite |
For Quality
| Task | Recommended Model |
|---|---|
| Creative writing | gpt-5, claude-opus-4.5 |
| Technical content | gpt-4o, claude-sonnet-4.5 |
| Code generation | gpt-5.1-codex, qwen3-coder |
| Research | o1, deepseek-r1 |
For Speed
| Task | Recommended Model |
|---|---|
| Quick responses | gpt-4o-mini, claude-haiku-4.5 |
| Batch processing | gemini-2.5-flash, grok-4-fast |
Troubleshooting Configuration
| Issue | Solution |
|---|---|
| API key not working | Verify key is correct and has credits |
| Model not available | Check if model is supported by your provider |
| Slow responses | Try a faster model (mini, flash, haiku) |
| Responses too short | Increase Max Tokens setting |
| Responses too varied | Lower Temperature setting |
Keep your API keys secure. Never share them or commit them to version control.
Multiple Providers
You can switch between providers at any time:
- Open SheetMagic Settings
- Select a different provider
- Enter the new API key
- Your formulas will automatically use the new provider
Tip: Use OpenRouter or Straico to access multiple providers with a single API key.