Models
Pick the right trade-off between quality, speed and cost. Prices are per 1M tokens.
- Default
Prism Pro
Best all-round model for writing, analysis and everyday work.
- Context window
- 200k tokens
- Time to first token
- 820 ms
- Input
- $3.00 / 1M
- Output
- $15.00 / 1M
- Vision
- Tools
- JSON mode
- Long context
10,907 requests in the last 30 days
Prism Fast
Near-instant answers for classification, rewriting and chat.
- Context window
- 128k tokens
- Time to first token
- 240 ms
- Input
- $0.30 / 1M
- Output
- $1.20 / 1M
- Tools
- JSON mode
20,010 requests in the last 30 days
Prism Reason
Thinks step by step through planning, math and hard trade-offs.
- Context window
- 200k tokens
- Time to first token
- 2.9 s
- Input
- $12.00 / 1M
- Output
- $60.00 / 1M
- Reasoning
- Tools
- Long context
1,486 requests in the last 30 days
Prism Code
Tuned for reading repositories, refactors, tests and reviews.
- Context window
- 256k tokens
- Time to first token
- 610 ms
- Input
- $2.00 / 1M
- Output
- $10.00 / 1M
- Code
- Tools
- JSON mode
- Long context
3,448 requests in the last 30 days
Prism Vision
Reads screenshots, charts, PDFs and handwritten notes.
- Context window
- 128k tokens
- Time to first token
- 1.4 s
- Input
- $2.50 / 1M
- Output
- $10.00 / 1M
- Vision
- JSON mode
1,223 requests in the last 30 days