Models & pricing / deepseek-v4-pro · DeepSeek
DeepSeek

deepseek-v4-pro API

DeepSeek-V4-Pro is a text model in the DeepSeek V4 family, suitable for evaluating reasoning over long material, coding and tool-assisted text tasks. XiuAI uses deepseek-v4-pro; its input modalities differ from the current image-capable Flash entry.[1][4]

deepseek-v4-pro

Model overview

Published provider specifications. Available features depend on the current API route.

API routes
OpenAI-compatible API

Long material and text-based tool tasks

Evaluate Pro for cross-document questions, code analysis or text tasks requiring several tool calls. Define the source boundaries, goal and checkable completion criteria; a large context does not eliminate retrieval, chunking or result review.[1][3]

How it differs from the current Flash entry

The official model list labels Pro as text input and text output; the current Flash also lists image input. Do not transfer Flash’s vision capability to Pro when a task requires raw screenshots or images. Compare their text behavior with the same prompt, material and output requirements.[1]

Vendor protocol metadata also distinguishes how models handle system-prompt updates. Similar message structures do not remove the need to check effective behavior during migration; one compatibility API does not establish behavior across every route.[1]

Long context is not a fixed task price

The current pricing panel distinguishes input, output and cache conditions. A long task can span several requests, with tool results becoming later input. Reconcile overall cost from actual usage and current catalog rates rather than extrapolating a short answer into a complete engineering-task budget.[4][3]

API access

Using the exact Pro identifier on XiuAI

Call deepseek-v4-pro through the OpenAI-compatible protocol listed in the current catalog, using your existing key and balance. Confirm a basic response before adding reasoning or tool parameters. DeepSeek-hosted tools, file services and account limits do not establish equivalent XiuAI services.[4][2]

Send a request using your existing API key.

POST /v1/chat/completions
curl 'https://router-api.xiu.ai/v1/chat/completions' \
  -H "Authorization: Bearer $XIUAI_API_KEY" \
  -H 'Content-Type: application/json' \
  --data-raw '{
  "model": "deepseek-v4-pro",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ]
}'
Read the API documentation

Frequently asked questions

Does Pro support image input?

The current official list labels deepseek-v4-pro as text input. For image understanding, review the current Flash specifications and XiuAI access; membership in the V4 family does not imply identical modalities.[1]

Do vendor-hosted tools come with model access?

Tool-calling support does not automatically include hosted tools. Model tool calls, application-side execution and vendor-hosted search or file services are separate layers. Use fields supported by the current API and tools integrated into your application.[3][4]

How should I compare Pro and Flash costs?

Keep prompts, source material, cache conditions and output requirements comparable, then compare current rates and actual usage. A name or one input rate does not establish total cost for completing the same task.[4]

Related models

DeepSeek V4.1 Flash

Sources and updates

Specifications reference provider documentation; prices and availability come from the current XiuAI catalog.

Content reviewed: 2026-11-16

[1] DeepSeek model list documentation[2] DeepSeek access and model mapping[3] DeepSeek API platform[4] Current XiuAI model and pricing dataTool callingStructured output