Simple, transparent pricing
Choose the plan that's right for you. All plans include access to our unified API.
FAQ
How do credits work?
Credits are the currency you use to make API requests. Each model has separate multipliers for input and output tokens.
Per-Token Pricing
Credits are calculated using separate input and output multipliers: (input × inputMult) + (output × outputMult). Output tokens typically cost more than input tokens.
Daily Reset
Your credits reset every 24 hours. Unused credits don't roll over to the next day, so make the most of your daily allowance.
Fixed-Cost Models
Image, video, audio, and embedding models have a fixed credit cost per request, regardless of input/output size.
Credit Calculation Formula
credits = (inputTokens × inputMultiplier) + (outputTokens × outputMultiplier)- •Lightweight models (mini, flash, nano) — input: 0.02×–0.03×, output: 0.08×–0.12×
- •Standard models (GPT-4o, GPT-5, Gemini Pro) — input: 0.25×–0.5×, output: 1×–2×
- •Premium models (Claude Opus, o1) — input: 1×–3×, output: 5×–15×
- •Fixed-cost models — Image gen: 1K–75K, Video gen: 5K–15K, TTS: 75–250, Embeddings: 50
- •Prompt caching — Cached input tokens are charged at a reduced rate (10%–50% of normal input cost)