DeepSeek V4 Flash

70% below official

DeepSeek's fast, budget V4 model for everyday chat, coding help and batch jobs — one of the most cost-effective capable models for running coding agents at volume.

Get an API keyQuickstart
Input
$0.09
Output
$0.18
Cached input
$0.03
Context
1M
Family
DeepSeek

Pricing

USD per 1M tokens, charged from your prepaid balance. How billing works

Input$0.09Official $0.30
Output$0.18Official $1.20
Cached input$0.03Official $0.006

Estimated cost per request $0.00027

Official comparison methodology & sources

For the selected input length, each rate uses the higher of domestic and overseas official standard pay-as-you-go prices. Domestic rates are converted to USD. Savings use the lower input/output discount; storage, tool fees and batch discounts are excluded. Image, video and audio models are compared at the same model version and unit; our video price is the same at every resolution, so the comparison uses the official price at the lowest resolution the model supports. Only models with a published official price are shown.

Exchange rate: 1 USD = 6.692782 CNY · 2026-10-12 · ECB ↗

Official price sources: 智谱 GLM CN ↗ Global ↗ · DeepSeek CN ↗ Global ↗ · 豆包 / Doubao · Kimi CN ↗ Global ↗ · MiniMax CN ↗ Global ↗ · OpenAI Global ↗ · xAI Global ↗ · BytePlus · 阿里云百炼 / Model Studio Global ↗ · Krea Global ↗ · Stability AI Global ↗ · Kling Global ↗ · Vidu Global ↗ · MiniMax 视频 Global ↗ · LTX Global ↗ · Prices checked 2026-10-18

API

POST /v1/chat/completions

OpenAI-compatible. Use your BigModel.io key with any OpenAI SDK by changing the base URL.

Request example

Set BIGMODEL_API_KEY in your terminal to your API key, then run one of these. The Python and Node.js samples use the official OpenAI SDK.

curl https://bigmodel.io/v1/chat/completions \
  -H "Authorization: Bearer $BIGMODEL_API_KEY" \
  -H "Content-Type: application/json" \
  --data '{
  "model": "deepseek-v4-flash",
  "messages": [
    {
      "role": "user",
      "content": "Hello!"
    }
  ]
}'

Create an API key for this model →

Manage API keys →

Use in Claude Code, Cursor and other tools

Set the model to deepseek-v4-flash after following the guide for your tool.

On BigModel.io

Model ID
deepseek-v4-flash
Platform context limit
1,000,000 tokens
Platform output limit
64,000 tokens
Inputs
Text
Outputs
Text
Developer
DeepSeek