V4-Pro is out of preview. The Flash build quietly learned to speak Codex. DeepSeek is betting price-performance is the wedge into agentic coding.
DeepSeek’s V4-Pro-0813 reached general availability on August 13, with a new price list effective August 16 at 16:00 UTC.
The V4-Flash 0731 build kept its architecture but was fully retrained in post-training — and added native Codex support alongside the Responses API.
DeepSeek’s changelog claims agent capabilities "far exceeding V4-Pro-Preview." Third-party numbers point the same way: 82.7 on Terminal Bench 2.1, 54.2 on NL2Repo.
The spec sheet — 284B total parameters, 13B active, up to 384,000-token outputs — targets long-horizon coding agents rather than chat.
Claude Code integrations on the 0731 build are already demoed in the wild.
V4 Pro moves from preview to general availability with general access across API and cloud partners, published pricing, and enterprise support commitments — the boring-but-critical milestones that turn a benchmark story into a procurement option.
The headline remains cost: GA pricing undercuts incumbent frontier models by a wide margin while claiming comparable performance on reasoning and coding suites. Enterprise buyers we spoke with describe running the same eval battery on both and finding the gap small enough that price decides.
Data governance is the one that stalls deals: where inference runs, what jurisdiction governs it, and how enterprise data controls map onto the new stack. Watch for compliance certifications and regional hosting announcements — those will unlock the buyers the benchmarks already convinced.
This story was reported from primary materials: official documentation, on-the-record statements and data we could independently check. Numbers were re-verified against original sources rather than secondary aggregations, and analyst commentary is labeled as commentary — not reporting. Where we could not confirm a detail, we said so in the text. Corrections update the article in place with the change noted at the top.
Three signals matter from here: whether early-adopter sentiment survives the honeymoon window, whether pricing converts attention into durable revenue, and how competitors answer — in this category, responses arrive in weeks, not quarters. Second-day stories are usually bigger than launch-day headlines; we keep this article updated as the picture firms up.
Zoom out and this story is one data point in a pattern: capability announcements, immediate commoditization, and a market that reprices in weeks what used to take years. For buyers, the practical lesson is to negotiate shorter contracts and keep exit paths open. For builders, it is that distribution and trust now matter more than model access — the raw capability is becoming the cheapest part of the stack.
If this story affects your tooling decisions, the actionable step is small: list the two workflows you run most often and check whether this change improves, ignores or complicates them. Most coverage, including ours, is written for everyone — your decision only has to be right for you.
The target isn’t chat. It’s the agent that ships code.
“Significantly enhanced agent capabilities, with benchmark results far exceeding V4-Pro-Preview.”
“V4-Flash holds up unusually well on bounded coding — 284B total parameters, 13B active, up to 384,000-token output.”
