DeepSeek V4 Flash Vision Exp - MixRoute API

DeepSeek V4 Flash Vision Exp is the experimental vision-capable V4 Flash route for text-and-image reasoning and multimodal agent workflows. Call deepseek-v4-flash-vision-exp through MixRoute using the endpoint below.

Key capabilities

Vision input - Accepts text and image content
1M context window - Supports long multimodal sessions
384K maximum output - Supports extended reasoning and generation
Tools and Responses compatibility - Supports agent-oriented workflows

Quick example

cURL

curl "https://api.mixroute.ai/v1/chat/completions" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "deepseek-v4-flash-vision-exp",
  "messages": [{
    "role": "user",
    "content": [
      {"type": "text", "text": "Describe the image."},
      {"type": "image_url", "image_url": {"url": "https://example.com/image.jpg"}}
    ]
  }],
  "max_tokens": 256
}'

Parameters

Parameter	Type	Required	Description
`model`	string	Yes	Model ID. Must be `deepseek-v4-flash-vision-exp`.
`messages`	array	Yes	Conversation messages with role and content.
`image_url`	object	No	Public URL or supported encoded image input.
`max_tokens`	integer	No	Maximum generated tokens.
`reasoning_effort`	string	No	Reasoning level when supported.
`tools`	array	No	OpenAI-compatible tool definitions.

Vendor reference

DeepSeek models and pricing