The public rate card
Every model on sale, with its four token prices per million tokens, ordered by vendor family and model id. Billing follows exactly what the upstream reports.
| Display name | Model id to send | Model vendor | Input | Output | Cache-write rate | Cache-read rate |
|---|---|---|---|---|---|---|
| Claude Fable 5 | claude-fable-5 | Anthropic | $4.35 | $21.74 | $5.35 | $0.3333 |
| Claude Fable 5.1 | claude-fable-5-1 | Anthropic | $4.38 | $21.88 | $5.39 | $0.3333 |
| Claude Haiku 4.5 | claude-haiku-4-5-20251001 | Anthropic | $0.4167 | $1.97 | $0.4861 | $0.0333 |
| Claude Opus 5 | claude-opus-5 | Anthropic | $1.97 | $9.83 | $2.42 | $0.1222 |
| Claude Opus 5.5 | claude-opus-5-5 | Anthropic | $1.79 | $8.46 | $2.11 | $0.0819 |
| Claude Sonnet 5 | claude-sonnet-5 | Anthropic | $1.22 | $6.14 | $1.54 | $0.1069 |
| Claude Sonnet 5.5 | claude-sonnet-5-5 | Anthropic | $1.25 | $6.28 | $1.57 | $0.1097 |
| DeepSeek V4 Flash | deepseek-v4-flash | DeepSeek | $0.2917 | $1.18 | $0.00 | $0.006944 |
| DeepSeek V4 Pro | deepseek-v4-pro | DeepSeek | $1.44 | $4.32 | $0.00 | $0.0486 |
| Gemini 2.5 Flash | gemini-2.5-flash | $0.2361 | $1.89 | $0.2361 | $0.0236 | |
| Gemini 3.1 Pro | gemini-3.1-pro | $1.47 | $8.56 | $1.47 | $0.1528 | |
| Gemini 3.8 Flash | gemini-3.8-flash | $1.35 | $6.10 | $1.35 | $0.1347 | |
| Gemini 4 Argon | gemini-4-argon | $1.72 | $8.28 | $1.72 | $0.0861 | |
| Kimi K3 | kimi-k3 | Moonshot Kimi series | $1.31 | $6.54 | $0.00 | $0.1306 |
| GPT-5.6 Sol | gpt-5.6-sol | OpenAI | $0.4306 | $2.42 | $0.00 | $0.0347 |
| GPT-6 Astra | gpt-6-astra | OpenAI | $0.8333 | $4.19 | $0.00 | $0.0833 |
| GPT-6.1 Sol | gpt-6.1-sol | OpenAI | $0.1528 | $0.7639 | $0.00 | $0.0306 |
| Qwen3.7 Plus | qwen3.7-plus | Alibaba Qwen series | $0.1667 | $0.4861 | $0.00 | $0.0486 |
| Qwen3.8 Max | qwen3.8-max | Alibaba Qwen series | $0.8611 | $2.58 | $0.00 | $0.1028 |
| Hunyuan 4 | hy4-preview | Tencent Hunyuan series | $0.4167 | $1.32 | $0.00 | $0.0194 |
| Grok 4.6 | grok-4.6 | xAI Grok series | $0.6667 | $2.01 | $0.00 | $0.1667 |
| Grok 4.7 | grok-4.7 | xAI Grok series | $0.625 | $1.88 | $0.00 | $0.1528 |
| GLM-5.3 | glm-5.3 | Zhipu GLM series | $0.6528 | $1.97 | $0.00 | $0.1667 |
How the prices apply
- Unit: $ per million tokens. Non-Chinese pages show an approximate USD figure at 7.2 CNY per dollar; charges are made in CNY.
- Cache writes and cache reads use their own prices; cached input reported by the upstream is taken out of the input count, so nothing is billed twice.
- If the upstream answers successfully but reports no usage, the call is estimated at 2,000 input and 2,000 output tokens and marked as an estimate in Usage.
- Accounts you refer earn you 19% of their top-ups, and 5% one level further down.
- Unused balance can be refunded within 15 days of the top-up.