Claude Models
Complete list of Anthropic Claude models supported by SheetMagic including Opus, Sonnet, and Haiku
Overview
Claude by Anthropic is known for its strong reasoning capabilities, safety features, and long context windows. SheetMagic supports all major Claude model families.
API Key: Get your API key at console.anthropic.com/settings/keys
Chat Models
Claude 4.5 Series (Latest)
| Model ID | Name | Description |
|---|---|---|
claude-opus-4-5-20251101 | Claude Opus 4.5 | Most capable model |
claude-sonnet-4-5-20250929 | Claude Sonnet 4.5 | Balanced performance |
claude-haiku-4-5-20251001 | Claude Haiku 4.5 | Fast, efficient |
Claude 4.1 Series
| Model ID | Name | Description |
|---|---|---|
claude-opus-4-1-20250805 | Claude Opus 4.1 | High capability |
Claude 4 Series
| Model ID | Name | Description |
|---|---|---|
claude-opus-4-20250514 | Claude Opus 4 | Advanced reasoning |
claude-sonnet-4-20250514 | Claude Sonnet 4 | Balanced model |
Claude 3.7 Series
| Model ID | Name | Description |
|---|---|---|
claude-3-7-sonnet-20250219 | Claude Sonnet 3.7 | Strong performance |
Claude 3.5 Series
| Model ID | Name | Description |
|---|---|---|
claude-3-5-haiku-20241022 | Claude Haiku 3.5 | Fast, cost-effective |
Claude 3 Series
| Model ID | Name | Description |
|---|---|---|
claude-3-opus-20240229 | Claude Opus 3 | Original flagship |
claude-3-haiku-20240307 | Claude Haiku 3 | Fastest option |
Model Tiers
Opus (Most Capable)
- Best for: Complex analysis, research, creative writing
- Context: 200K tokens
- Strengths: Nuanced reasoning, detailed outputs
Sonnet (Balanced)
- Best for: General tasks, content generation, coding
- Context: 200K tokens
- Strengths: Balance of quality and speed
Haiku (Fastest)
- Best for: Quick tasks, simple queries, high-volume operations
- Context: 200K tokens
- Strengths: Speed, cost-efficiency
Model Capabilities
| Capability | Opus | Sonnet | Haiku |
|---|---|---|---|
| Complex reasoning | Excellent | Good | Basic |
| Code generation | Excellent | Excellent | Good |
| Creative writing | Excellent | Good | Basic |
| Summarization | Excellent | Excellent | Good |
| Translation | Excellent | Excellent | Good |
| Speed | Slower | Medium | Fast |
| Cost | Higher | Medium | Lower |
Usage Examples
Basic Text Generation
=AITEXT("Write a detailed analysis of renewable energy trends")Code Generation
=AITEXT("Write a Python function to calculate compound interest with monthly contributions")Content Summarization
=AITEXT("Summarize this article in 3 bullet points: " & A1)Translation
=AITEXT("Translate to Spanish: " & A1)Key Features
Long Context Window
Claude models support up to 200K tokens of context, allowing you to:
- Process entire documents
- Analyze long-form content
- Maintain conversation history
Safety & Alignment
Claude is designed to be:
- Helpful and harmless
- Honest about limitations
- Resistant to jailbreaks
Constitutional AI
Claude uses Constitutional AI training, making it:
- More reliable for business use
- Better at following guidelines
- Less prone to harmful outputs
Configuration
Temperature
Controls creativity/randomness (0-1):
0: Deterministic responses0.5: Balanced (default)1: More creative
Max Tokens
Controls maximum response length. Claude supports large outputs up to the context limit.
Best Practices
1. Choose the Right Model
| Use Case | Recommended |
|---|---|
| Complex analysis | Opus 4.5 |
| Daily tasks | Sonnet 4.5 |
| High-volume, simple tasks | Haiku 4.5 |
| Budget-conscious | Haiku 3.5 |
2. Leverage Long Context
Claude excels at processing long documents:
=AITEXT("Analyze this entire document and provide key insights: " & VISIT(A1))3. Use Clear Instructions
Claude responds well to structured prompts:
=AITEXT("Task: Write a product description
Product: " & A1 & "
Tone: Professional
Length: 100 words
Include: Features, benefits, call-to-action")4. Batch Processing
For high-volume tasks, use Haiku for cost efficiency:
// Configure SheetMagic to use claude-3-5-haiku
// Then process hundreds of cells efficientlyPricing Considerations
Claude pricing is based on:
- Input tokens (your prompt)
- Output tokens (response)
- Model tier (Opus > Sonnet > Haiku)
Haiku 3.5 is significantly cheaper than Opus, making it ideal for bulk operations where top-tier quality isn't required.
Limitations
- No image generation: Claude doesn't create images
- No speech synthesis: No TTS capabilities
- No video generation: No video creation
For these features, use OpenAI, Gemini, or other providers.