DeepSeek V5 status tracker
Still zero from DeepSeek — and the V4 line already moved to 4.1
As of 2026-10-07, neither DeepSeek's official changelog nor its news page mentions V5; the latest entry is still V4.1-Flash from 2026-09-10. After V4 Pro went GA on 2026-08-13 and the 2026-08-17 price rise landed, rumours of a September V5 at Mythos level spread, but DeepSeek has said nothing about V5. This page tracks only what can be verified.
Updated 2026-10-07
Current status at a glance
V5's current phase
In mid-August the web circulated 'September release, Mythos-class performance.' Zero official word from DeepSeek.
V4 Pro official GA
1M context, 384K max output, with native Responses API and Codex support.
API price hike effective
Peak/off-peak pricing landed: V4-Pro peak output ¥27 per million tokens (+350%), cached input up as much as 1100%.
The V4 series on QCode
As of 2026-10-07, QCode serves deepseek-v4.1-flash, deepseek-v4-flash and deepseek-v4-pro. DeepSeek's pricing page notes that the legacy name deepseek-v4-flash points to a retired model and its requests are served by DeepSeek-V4.1-Flash and billed at the Flash price — on QCode, new integrations should use deepseek-v4.1-flash. Billed per token; per-model prices are on /models.
Where the V5 rumor comes from
The source is industry interpretation after the 08-17 price hike took effect: in the price-change announcement, DeepSeek mentioned it 'will develop stronger large models,' and media stitched that together with supply-chain whispers into 'V5 in September, benchmarked against Mythos 5.' Note: the announcement itself never says 'V5' — 'September' and 'Mythos-class' are both secondhand extrapolation.
The V4 line is still iterating fast
While everyone waits for V5, the V4 line did not stop: V4 Pro GA on 2026-08-13 (officially stated: Terminal-Bench 2.1 87.9); multimodal V4-Flash-Vision-Exp on 2026-08-21; V4.1-Flash released 2026-09-10, which the note calls “the smallest model in our new architecture family” with native visual understanding — and on the same day deepseek-v4-flash and deepseek-v4-flash-vision-exp were retired (pricing footnote (1): requests on those names are served by V4.1-Flash at the Flash price). The changelog entry of the same day states that API service for DeepSeek V4 Pro continues after 2026-09-14 with billing unchanged (the release note's rerouting sentence is quoted in the timeline). As of 2026-10-07 the latest entry in both the changelog and the news page is still 2026-09-10, and V5 still has zero official information.
Timeline
2026-08-06: the developer console previews a broad API price rise, “with a large increase”.
2026-08-13: V4 Pro GA plus the pricing plan — peak/off-peak, peaks 09:00–12:00 and 14:00–18:00 Beijing time.
2026-09-10: V4.1-Flash released. The release note says: “Starting at 04:00 UTC on Sept 14, 2026, all deepseek-v4-pro requests will route to V4.1-Flash at V4.1-Flash rates. This will continue until V4.1-Pro launches.” The changelog entry of the same day says: “we have decided to continue providing API services for DeepSeek V4 Pro after September 14, 2026, with the billing method remaining unchanged.” As of 2026-10-07 both pages still carry these sentences, and the pricing page still lists deepseek-v4-pro (DeepSeek-V4-Pro-0813).
Confirmed vs rumors
Confirmed
V4 Pro GA 2026-08-13, peak/off-peak rise effective 2026-08-17, V4-Flash-Vision-Exp on 2026-08-21, V4.1-Flash on 2026-09-10 retiring deepseek-v4-flash and deepseek-v4-flash-vision-exp, and the changelog's 2026-09-10 entry officially states that V4 Pro API service continues after September 14, 2026 with billing unchanged — all backed by official notes or the pricing footnote. V5 has none of it.
Rumors
'V5 in September' and 'Mythos-class' have no official source — they're amplified readings of the phrase 'develop stronger large models' in the price-hike announcement. Even the name V5 itself is unconfirmed.
Wait for V5 or use V4
Wait for V5
Fits: non-urgent needs. Risks: no date, no specs — and a new generation will most likely inherit the new pricing regime.
Use V4 Pro / Flash
As of 2026-10-07 the official pricing page lists two models: deepseek-flash (version DeepSeek-V4.1-Flash) and deepseek-v4-pro (version DeepSeek-V4-Pro-0813). Per 1M tokens, deepseek-flash input (cache miss) is $0.15 off-peak / $0.3 peak and output $0.6 / $1.2; deepseek-v4-pro input is $0.66 / $1.32 and output $1.98 / $3.96. The page says: “Off-peak rates are half of the peak rates. Peak hours are 01:00 - 04:00 and 06:00 - 10:00 UTC, Monday through Friday, excluding Chinese public holidays.”
How to follow along
DeepSeek only recognizes two channels: the official docs site (api-docs.deepseek.com) and the official WeChat account. If V5 exists, it will show up in one of those two places first. This page tracks in sync.
On QCode
There is no callable V5. On QCode you can call deepseek-v4.1-flash, deepseek-v4-flash and deepseek-v4-pro, billed per token, with per-model prices on /models. DeepSeek's pricing page notes that requests to the legacy name deepseek-v4-flash are served by DeepSeek-V4.1-Flash — build new integrations on deepseek-v4.1-flash.
FAQ
Has DeepSeek V5 been released?
No. As of 2026-10-07 the latest entry in DeepSeek's official changelog and news page is still V4.1-Flash from 2026-09-10, and the pricing page lists only deepseek-flash and deepseek-v4-pro — none of the three mentions V5.
What's the basis for the V5 rumor?
A single official phrase: the price-hike announcement mentioned 'will develop stronger large models.' Everything beyond that is speculation.
How much did V4 prices rise?
V4-Pro peak output is ¥27 per million tokens (about +350% over before); cached input rose as much as 1100%; V4-Flash rose up to 400%. Off-peak is half of peak.
Is DeepSeek still cheap after the hike?
It depends on how you use it: the official pricing page says off-peak rates are half of the peak rates, and cache-hit input has its own, lower price — so off-peak traffic with a high cache-hit rate costs the least.
How does V4 Pro GA differ from Preview?
The GA build (0813) significantly strengthens agentic capability, natively supports the Responses API and Codex, and scores 87.9 on TB2.1 by official terms.
What is the multimodal V4-Flash-Vision?
An experimental build launched 08-21: images are billed as tokens (up to 384 tokens per image), priced identically to the regular V4-Flash.
Sources
DeepSeek's pricing note (2026-08-13, reported by Xinhua Finance), Guancha (2026-08-17), Caixin (2026-08-21), Kaiyuan Securities (2026-08-06), KuaiKeji (2026-08-18, the V5 rumour); plus DeepSeek's news260910 release note and official changelog (first archived 2026-09-18) as well as the Models & Pricing page; the English and Chinese editions of all three were fetched and checked on 2026-10-07.
Until V5 arrives, V4 is in stock
deepseek-v4.1-flash, deepseek-v4-flash and deepseek-v4-pro are ready to call on QCode, billed per token — prices on /models.
Related reading
DeepSeek V4 Pro 0813
The GA build's specs and how to call it.
DeepSeek V4.1 Pro status tracker
The tracker for the note's only "until V4.1-Pro launches" line: no date, no id, no price.
DeepSeek V4 price increase (2026)
Scheduling tricks to save under peak/off-peak pricing.
Not affiliated with DeepSeek. All V5-related content is rumor and is labeled as such in the text. Last checked 2026-10-07; official pages take precedence, and model availability is as listed on /models.