DeepSeek's official X account says the V4 API discount is now permanent, and the current pricing page still matches the lower Flash and Pro rates

Checked on July 1, 2026: DeepSeek's official X account published a permanent-discount update, and the current official Models & Pricing page still shows the lower DeepSeek V4 Flash and Pro token rates together with the existing 1M-context and 384K max-output contract.

Accepted official-source monitoring note

Today's accepted item stays DeepSeek-first and uses the newest official @deepseek_ai post we could verify safely in search plus DeepSeek's own pricing docs. The direct X timeline still was not reliably crawlable in this run, but DeepSeek's English homepage still anchors the official X account and search-indexed results exposed the official status URL cleanly enough to confirm the headline.

What we verified on July 1, 2026

DeepSeek's official X account published a permanent-discount update and framed the price cut as an official DeepSeek API policy rather than a temporary teaser.
DeepSeek's current official Models & Pricing page still matches that lower-rate posture: deepseek-v4-flash is listed at $0.0028 cache-hit input, $0.14 cache-miss input, and $0.28 output per 1M tokens, while deepseek-v4-pro is listed at $0.003625, $0.435, and $0.87 respectively.
The same current pricing page still keeps the wider V4 contract intact: OpenAI and Anthropic base URLs, 1M context length, 384K max output, and the current Flash/Pro concurrency caps.
The pricing page also still warns that product prices may vary, so this publish treats the permanent-discount message as the current official posture, not as a promise that rates can never change again.
This is API pricing news, not local storefront inventory news: it does not create any new Coding Plan card, stock guarantee, or SUBSCRIPTION_PLANS change.

Why this is publishable

This is an official DeepSeek signal backed by two first-party surfaces: the official X status itself and the live pricing page that still reflects the discounted table.

It is newer and more specific than recycling the older V4 Preview homepage anchor.
It answers a real buyer question: whether the lower V4 rates still look like a temporary campaign or a current official baseline.
It stays inside official facts and avoids inventing any new inventory, roadmap, or SLA claim.
It stays distinct from the June 28 pricing-contract item because the headline here is discount permanence from the official X account, not just the numerical contract snapshot.

Why this matters for DeepSeek-first SEO pages

Buyer pages can now target DeepSeek permanent API discount intent with first-party backing instead of recycled community screenshots.
Comparison pages can separate current token rates from temporary sale language, which matters for teams budgeting long-running agent traffic.
Pricing support pages can explain that lower API rates do not automatically imply new local resale inventory on /pricing.

Rejected candidates today

Direct official X timeline as a browsable page: rejected because the live timeline still was not reliably crawlable in this run, even though the official status URL was verifiable through search indexing.
The DeepSeek English homepage V4 Preview anchor: official, but older and already represented heavily in existing launch coverage.
The official reasoning-model and chat-completions docs: accepted as stronger long-tail guide targets than as the single daily news headline.
The official status, GitHub, and Hugging Face surfaces: checked as backup official sources, but none provided a stronger current publish-safe headline than the X pricing update plus the live pricing table.

Editorial takeaway

The clearest official DeepSeek story today is a pricing posture check: DeepSeek's official X account says the API discount is permanent, and the current first-party pricing page still reflects the lower Flash and Pro rates. That is a stronger current headline than republishing an older product-launch anchor or forcing a narrower docs page into the news slot.