The DeepSeek V4 family from its maker's own API
DeepSeek's first-party Open Platform — the maker's own OpenAI-compatible API for the V4 family (Flash, Pro and the experimental vision Flash). Inference, retained inputs and the disk context cache are all in mainland China.
Call the 4 models on DeepSeek through Opper with one API key. Sign up without a credit card.
Every model we route through DeepSeek. Compare residency, zero data retention, training posture and price at a glance, with full data-handling detail per route below.
| Model | Context | Region | Zero data retention | Training | Input | Output | Cache read |
|---|---|---|---|---|---|---|---|
| 1M | China | Yes | $1.32 | $3.96 | $0.04 | ||
| 1M | China | Yes | $0.44 | $1.32 | $0.01 | ||
| 1M | China | Yes | $0.44 | $1.32 | $0.01 | ||
| 1M | China | Yes | $0.44 | $1.32 | $0.01 |
DeepSeek hosts on 1 route. Each route has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.
Content is retained: used for training by default. Trained on customer data by default. CN.
