Models & pricing
Model IDs
Every supported model id, its vendor, and what it's good at.
Model IDs
Pass any of these as the model field:
| claude-haiku-4-5 | Fast, low-cost chat |
| claude-sonnet-4-6 | Balanced reasoning |
| claude-opus-4-8 | Highest-capability reasoning |
| gpt-5-4-mini | Fast, low-cost chat |
| gpt-5-4 | Frontier chat / vision |
| gpt-5-5 | Flagship chat / tools |
| gemini-2-5-flash-lite | Cheapest high-volume chat |
| gemini-2-5-flash | Fast, low-cost chat |
| gemini-2-5-pro | Long-context frontier |
Multipliers
Every model debits your balance at its own rate: charged = real_tokens × multiplier. Lighter models (Haiku, GPT mini, Flash-Lite) sit around a 1× multiplier; top-end reasoning models cost more per token. Each model's exact multiplier is shown in the dashboard.