Noesion Radarnoesion.ai
Release2026-08-12

DeepSeek began rolling out its V4-Pro-0813 model through its own programming interface and chat app, priced at $0.435 per million input tokens and $0.87 per million output tokens. Early benchmark figures circulating on WeChat and from third parties have not been independently verified. DeepSeek runs on roughly 20,000 Nvidia H100 chips and has seen response speeds slow under heavy demand.

What changed

DeepSeek's top-tier option was its April V4 preview, with the cheaper V4-Flash released weeks earlier.

What it unlocks

Running a large frontier-class model at roughly a third of the per-token cost of comparable rivals, via DeepSeek's API or chat app.

  • $0.435 per 1M input tokens
  • $0.87 per 1M output tokens
  • 1.6T parameters, 49B active
  • 1M token context

What you need to act on it

  • a DeepSeek API key or account
  • willingness to send data to a China-based provider

Send this to someone who needs it

Shares the story and its sources. Nothing about you.

What does this mean for your job?

This is the story as everyone gets it. Once a week we send you the version written for your role — what changed, why it matters for the work you actually do, and one thing to try. Free while we tune it.