DeepSeek V4 Pro API: pricing, base URL and quick setup

Specifications and pricing

Prices synced with platform billing (USD per 1M tokens).

Quick setup

curl https://a2agent.me/v1/chat/completions \
  -H "content-type: application/json" \
  -H "authorization: Bearer sk-your-a2agent-key" \
  -d '{
    "model": "deepseek-v4-pro",
    "messages": [{"role": "user", "content": "Reply with OK"}]
  }'

Using it from a coding client

Reasoning-heavy sessions in Claude Code and Codex CLI are where the Pro rate earns its keep. If a session is mostly file reading and small edits, Flash does the same job on the same window for a third of the money.

• Run it in Claude Code
• Run it in Codex CLI

How it compares within its family

DeepSeek sells two V4 tiers here. Both carry a 1M context and both are tagged for reasoning, so the split is cost against capability rather than a feature list. V4 Pro lists at $0.435 per million input tokens and $0.87 per million output. V4 Flash lists at $0.14 and $0.28. That is roughly three times the rate for the same window and the same tags. The rule of thumb falls out of the numbers: reach for Pro when a wrong answer costs more than the tokens it took to get it, and drop to Flash on bulk work where you would rather run the job three times and compare the answers. Nothing in the catalogue data separates the two on context, so a prompt that fits Flash fits Pro as well.

Other models on this platform

• DeepSeek V4 Flash1M

Questions