Simple, Transparent Pricing
Subscribe monthly for the best rate, or buy once for full flexibility. Per-token pricing across 200+ AI models, billed in USD, CNY, or IDR. No hidden fees, no minimum commitment.
The live per-model pricing table loads in your browser with currency conversion to USD / CNY / IDR based on your locale. If you're seeing this text, JavaScript is loading — the interactive table will appear in a moment.
Sample model prices (USD per 1M tokens)
List prices for popular models, pulled live from the catalog. See the full interactive table above for all 200+ models and currency conversion to CNY / IDR.
| Model | Input / 1M | Output / 1M |
|---|---|---|
| gpt-5.5 | $2.75 | $16.50 |
| claude-opus-4.6-thinking | $5.00 | $25.00 |
| claude-sonnet-4-5-20250929 | $3.00 | $15.00 |
| claude-haiku-4.5 | $1.00 | $5.00 |
| gemini-3.8-flash | $0.750 | $3.75 |
| siliconflow/deepseek-v3.2 | $0.287 | $0.431 |
| siliconflow/deepseek-r1-0528 | $0.574 | $2.29 |
| qwen3.5-omni-flash-2026-03-15 | $0.299 | $2.44 |
| grok-4.7 | $2.00 | $6.00 |
How pricing works
- • Every model has its own per-token rate. Input tokens and output tokens are billed independently.
- • Subscribe monthly (Developer / Pro / Business) for the best per-token rate, with premium tokens bundled in each renewal.
- • Prefer flexibility? Buy the same packs once at a small markup, or top up your wallet and pay as you go.
- • Failed requests due to provider outages are refunded automatically.
- • Multi-currency display: prices render in USD, CNY, or IDR matching your locale at the admin-set exchange rate.
Cut your bill with auto-routing
Replace a specific model name with celedog/auto-cheapest and Celedog picks the cheapest model that satisfies your quality bar per request — often a 20–40% cost reduction without sacrificing output quality. See the auto-routing tutorial for details.