APIs
Call lite, balanced, or frontier on chat completions. Call midium on embeddings and on System One. Prices below are per million tokens for that API. Dedicated capacity is a flat rate for the same names.
| API | Model name | Input / M | Output / M |
|---|---|---|---|
| LiteGemma 4 E2B | lite | $0.11 | $0.55 |
| BalancedGemma 4 26B | balanced | $0.28 | $1.12 |
| FrontierOrnith 1.5 122B | frontier | $1.05 | $5.25 |
| Embeddings | midium | $0.012 | $0.036 |
| System OneDecider 4B | midium | $0.032 | $0.00 |
Dedicated capacity tiers
Starter · Mac Mini 48 GB
- Throughput
- 3–6 billion tokens a month
- Team size
- 1–5 team members
- Model class
- Lite
Dedicated capacity · $444/mo
Hub license · $200/mo
Standard · Mac Mini 64 GB
- Throughput
- 6–9 billion tokens a month
- Team size
- 6 or more team members
- Model class
- Balanced
Dedicated capacity · $589/mo
Hub license · $200/mo
Max · Mac Studio 256 GB
- Throughput
- 9–42 billion tokens a month
- Team size
- Up to 30 engineers
- Model class
- Frontier
Includes Qwen3.8 Flash Next.
Dedicated capacity · $2,149/mo
Hub license · $500/mo
Ultra · Mac Studio 512 GB
- Throughput
- 66–84 billion tokens a month
- Team size
- 30 or more engineers
- Model class
- Frontier
Includes GLM 5.3 Flash.
Dedicated capacity · $3,400/mo
Hub license · $500/mo