Meldport
DeepSeekactive

DeepSeek V4 Flash 0731

meldport/deepseek-v4-flash-0731-alicn

DeepSeek V4 Flash 0731 is available through Meldport as alicn/deepseek-v4-flash-0731. Published capabilities: Text, Cache, Stream, Tools.

Get API access

MODEL OVERVIEW

About DeepSeek V4 Flash 0731

DeepSeek V4 Flash 0731 is available through Meldport as alicn/deepseek-v4-flash-0731. Published capabilities: Text, Cache, Stream, Tools.

Use the published model ID in requests to Meldport's API. Check the current pricing, context window, and supported capabilities on this page before sending production traffic. See the quickstart for a request example.

GOOD FIT FOR

  • Text generation, extraction, and automation
  • Tool calling and agent workflows

BEFORE YOU SHIP

  • Check capabilities, context limit, and price against your workload.
  • Test tool call parsing and error handling before production.

TECHNICAL SPECIFICATIONS

What this model supports

Model IDmeldport/deepseek-v4-flash-0731-alicn
PublisherDeepSeek
CapabilitiesText, Cache, Tools
Context window1M
Lifecycleactive
VerifiedNov 1, 2026

CURRENT RATES

Pricing

MetricRateUnit
Input$0.066per million tokens
Output$0.198per million tokens
Provider-reported cached input$0.0066per million tokens
Minimum charge$0.00001per request

Provider-reported cached input tokens use the displayed rate. Every request is recorded in the wallet ledger.

QUICKSTART

Call this model

The same endpoint works from any language.

Send POST /v1/chat/completions with an API key that has inference:invoke. Use meldport/deepseek-v4-flash-0731-alicn in the model field. Supply your conversation in messages and set stream: true for streamed output. The published context limit is 1M tokens.

This model publishes tool calling. Send function definitions in tools and return tool results in the conversation. Request fields and output limits · Streaming format · Errors and retry rules

curl https://api.meldport.com/v1/chat/completions \
  -H "Authorization: Bearer $MELDPORT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "meldport/deepseek-v4-flash-0731-alicn",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

COMMON QUESTIONS

Frequently asked questions

What is the API model ID for DeepSeek V4 Flash 0731?

Use meldport/deepseek-v4-flash-0731-alicn in the model field of your API request.

Which capabilities does DeepSeek V4 Flash 0731 support?

This model currently publishes Text, Cache, Tools support through Meldport.

How is DeepSeek V4 Flash 0731 priced?

Requests are billed per token. Input costs $0.066 per million tokens and output costs $0.198 per million tokens. Provider-reported cached input tokens cost $0.0066 per million tokens; Meldport response caching is bypassed. A minimum charge of $0.00001 applies per request. Charges are deducted from your wallet balance in real time.

Which endpoint runs DeepSeek V4 Flash 0731?

Send a request to POST https://api.meldport.com/v1/chat/completions with your Meldport API key and use meldport/deepseek-v4-flash-0731-alicn in the model field.

What context window does DeepSeek V4 Flash 0731 support?

The published context limit is 1M tokens.