Model comparison
| Model | Credit usage | Best for | Editing | Availability | Underlying model |
|---|---|---|---|---|---|
| Max | High | Complex reasoning, multi-document work, long summaries | Yes | All platforms (Plus+) | Claude Sonnet 4.6 |
| Fast | Medium | Space-level questions, deeper analysis, longer conversations | Yes | All platforms | Claude Haiku 4.5 |
| Core | Low | Everyday in-document queries, balanced performance | No | All platforms | ChatGPT Nano 5.1 |
| On-device | Free | Short rewrites, quick transformations, simple questions | No | Mac, iPad, iPhone (document level) | LLaMa 3.2 + local models |
When to choose each model
On-device
Use for quick summaries, sentence rewrites, and simple clarifications.- No credit usage
- Local processing
- Read-only conversations
Core
Use for most general-purpose, in-document questions.- Best credit efficiency among cloud models
- Good default for everyday use
- Read-only conversations
Fast
Use for space-level queries and quick edits.- Good balance of speed and quality
- Supports editing on supported platforms
- Works well for short to medium editing tasks
Max
Use for complex reasoning and high-quality editing.- Best output quality for large or structured edits
- Best for advanced multi-document analysis
- Supports editing on supported platforms
Model availability by platform
| Platform | On-device | Core | Fast | Max | Editing |
|---|---|---|---|---|---|
| macOS | ✓ | ✓ | ✓ | ✓ (Plus+) | ✓ |
| iOS | ✓ | ✓ | ✓ | ✓ (Plus+) | ✓ |
| Windows | – | ✓ | ✓ | ✓ (Plus+) | – |
| Web | – | ✓ | ✓ | ✓ (Plus+) | – |
Related Articles
Using Craft Assistant
Learn how to start conversations and work with the Assistant
Editing with Craft Assistant
See what editing can do and how modes work
Credits
Understand credit usage and purchase options