DeepSeek V4 Pro 0813 API

What is DeepSeek V4 Pro 0813 API in NextModel?

deepseek/deepseek-v4-pro-0813 is the date-pinned DeepSeek V4 Pro snapshot, listed at $1.33 per million input and $3.98 per million output tokens. The rolling deepseek/deepseek-v4-pro id sits at $1.65 / $3.31: cheaper on output, more expensive on input. If your traffic is output-heavy, the rolling id wins; the pin buys reproducibility, not verified extra quality. Both ids list a 1,000,000-token context window.

Deepseek-V4-Pro-0813 via NextModel: 1M-token context; $1.33 / 1M tokens in / $3.98 / 1M tokens out per 1M tokens. Capabilities: streaming, tool, json.

Best use cases

• You need V4 Pro behavior pinned to the 0813 snapshot for reproducibility.
• Input-heavy workloads where the pin's cheaper $1.33 input rate helps.
• You accept $3.98 per million output tokens.

When not to use

• Output-heavy traffic: the rolling deepseek-v4-pro id is listed at $3.31 output versus $3.98 here.
• You need a context window beyond 1,000,000 tokens: that is both ids' published limit.
• You want cheap DeepSeek tokens: deepseek-v4-flash is $0.14 / $0.28.

OpenAI-compatible code example

curl https://api.nextmodel.app/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $NEXTMODEL_API_KEY" \
  -d '{"model":"deepseek/deepseek-v4-pro-0813","messages":[{"role":"user","content":"Reply with OK."}]}'

Try it in the playground

Similar alternatives

Compare Deepseek-V4-Pro-0813