DeepSeek V4 Pro is available through Anyfast via an OpenAI-compatible interface. It features thinking mode, function calling, and JSON mode support.Documentation Index
Fetch the complete documentation index at: https://docs.anyfast.ai/llms.txt
Use this file to discover all available pages before exploring further.
Key capabilities
- OpenAI-compatible — Works as a drop-in replacement with the OpenAI SDK
- Thinking mode — Supports reasoning with configurable effort levels
- Function calling — Call tools and functions natively
- JSON mode — Structured output via
response_format - Long context — Handles large documents and multi-turn conversations
- Streaming — Supports real-time token streaming via SSE
- Prompt caching — Automatic cache with hit/miss token reporting
Quick example
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
model | string | Yes | Must be deepseek-v4-pro |
messages | array | Yes | List of { role, content } objects. Supports system, user, assistant, tool roles. |
thinking | object | No | Enable/disable thinking: {"type": "enabled"} or {"type": "disabled"}. Optional reasoning_effort: high, max, low, medium. |
max_tokens | integer | No | Maximum tokens to generate |
temperature | float | No | 0–2. Controls randomness. Default: 1 |
stream | boolean | No | Enable SSE streaming. Default: false |
stream_options | object | No | {"include_usage": true} includes token usage in the final streaming chunk |
top_p | float | No | Nucleus sampling threshold. Default: 1 |
stop | string / array | No | Sequences that stop generation (up to 16) |
response_format | object | No | {"type": "json_object"} for JSON mode |
tools | array | No | List of function tools for function calling |
tool_choice | string / object | No | Control tool selection: none, auto, required, or specific function |
user_id | string | No | Custom user ID for content safety and KVCache isolation |
API Reference
View the interactive API playground for DeepSeek V4 Pro.