AI Model Guide

Migrating From DeepSeek V4 Pro to V4.1 Flash in 2026: The Retirement, the Reversal and the Checklist

Two official DeepSeek pages currently disagree about whether V4 Pro is being switched off this week. The right response does not depend on which one wins. This is the operational guide: what changed for callers, how to test the move, and when not to make it.

Distk Editorial Sep 2026 11 min read

On 10 September 2026 DeepSeek announced it was phasing out DeepSeek V4 Pro and that from 04:00 UTC on 14 September all deepseek-v4-pro requests would be routed to V4.1 Flash at V4.1 Flash rates until a V4.1 Pro launches. On 11 September the API pricing page added a footnote saying that in response to user demand V4 Pro service continues after 14 September with billing unchanged. Both pages are live. The planning posture is the same either way: V4 Pro has a published end-of-life intent and no date, V4.1 Flash is cheaper on every line and higher on most of DeepSeek's agentic benchmarks, and the migration is a config change plus an evaluation run. The caller-facing differences are native vision, a 1 to 100 reasoning effort dial, thinking on by default, a 2,500 concurrency limit, and Flash pricing. Stay on V4 Pro only where your own tests show it ahead on knowledge-heavy or long-context work and the cost difference is immaterial.

What Did DeepSeek Announce About V4 Pro in September 2026?

DeepSeek's V4.1 Flash announcement on 10 September 2026 stated three things about the existing lineup: V4 Flash and V4 Flash Vision Exp are retired with their legacy names temporarily routed to V4.1 Flash; tests by multiple parties put V4.1 Flash ahead of V4 Pro on performance, cost, speed and total runtime, so DeepSeek is phasing out V4 Pro; and starting 04:00 UTC on 14 September 2026 all deepseek-v4-pro requests would route to V4.1 Flash at V4.1 Flash rates until V4.1 Pro launches.

On 11 September 2026 the API pricing page, which still lists deepseek-v4-pro as a model, added a footnote: in response to user demand, DeepSeek has decided to continue providing API services for V4 Pro after 14 September with the billing method unchanged, and will provide further notice should there be any changes. Nothing on the announcement page was edited to match. At the time of writing, a reader of the announcement believes V4 Pro is being switched off this week and a reader of the pricing page believes it is not.

SourceDateWhat it says about V4 ProHow to weigh it in 2026
V4.1 Flash announcement10 September 2026Phasing out; all V4 Pro requests route to V4.1 Flash at Flash rates from 04:00 UTC 14 September until V4.1 Pro launchesStates the intent and the eventual destination.
API pricing page, footnote 211 September 2026V4 Pro API service continues after 14 September, billing unchanged, further notice if anything changesLater, more operational, and on the page that governs billing. Treat as current.

The reasonable synthesis is that V4 Pro is a model with an announced end of life and a paused switch-off. The August V4 Pro rate change was also announced with days of notice, so a team on V4 Pro in 2026 should assume the next notice will be short too. The V4.1 Flash overview covers what the replacement model is.

Why Should Most V4 Pro Callers Migrate to V4.1 Flash Anyway in 2026?

Because V4.1 Flash is cheaper on every line of the rate card, higher on most of DeepSeek's own agentic benchmarks, and has five times the concurrency, and because DeepSeek has told you where the traffic is eventually going. The pricing guide has the full rate comparison; the short version is that V4.1 Flash is about 7 times cheaper on cache hits, 4.4 times on cache-miss input and 3.3 times on output.

On DeepSeek's instruct benchmarks at maximum effort, V4.1 Flash beats V4 Pro on Terminal-Bench 2.1, 3.0 and 4.0, DeepSWE, AutomationBench, Agent's Last Exam, CyberGym, SEC-Bench Pro, HLE with tools and Codeforces. V4 Pro's base model retains an edge on knowledge-heavy rows such as SimpleQA-Verified (55.2 versus 42.3), MultiLoKo (50.9 versus 45.5) and LongBench-V2 (51.5 versus 45.2), and on GPQA Diamond (92.4 versus 90.9). The benchmark guide decodes the full table.

What Changes for Callers Moving to V4.1 Flash in 2026?

Most integrations will work by changing the model string, because both models sit behind the same OpenAI-format and Anthropic-format endpoints with the same feature list. The differences are in behaviour and limits, and each one needs a line in a migration plan.

DimensionDeepSeek V4 Pro 0813DeepSeek V4.1 FlashMigration action
Model namedeepseek-v4-prodeepseek-flashChange the string. Legacy deepseek-v4-flash names also resolve to V4.1 Flash.
Endpointsapi.deepseek.com (OpenAI format), api.deepseek.com/anthropicSameNone.
VisionNot supportedNative image inputRetire any separate vision model or OCR step; test image prompts.
Thinking modeNon-thinking and thinking, thinking defaultSameConfirm the mode your calls use; explicit is safer than default.
Reasoning effortNot described as a 1 to 100 dial on the pricing pageInteger 1 to 100; benchmarks at 100Set per workflow. Do not inherit 100 into production.
Context and output1M context; 384K max output1M context; 384K max outputNone.
Concurrency limit5002,500Raise client-side rate limiters if they were tuned to 500.
FeaturesJSON output, tool calls, Responses API, Anthropic API, chat prefix (beta), FIM non-thinking only (beta)SameNone.
Pricing0.044 / 1.32 / 3.96 USD per 1M at peak0.006 / 0.30 / 1.20 USD per 1M at peakRe-run the budget; expect a reduction, then verify token counts.
Prompt format (self-host only)V4 prompt formatV4.1 prompt format via encoding.py or deepseek-recipeAPI callers unaffected. Self-hosters see the self-host guide.

How Should a Team Migrate From V4 Pro to V4.1 Flash in 2026?

The same way any model migration should go in 2026: as a configuration change validated by an evaluation set, not as a search-and-replace followed by hope. The steps below assume a team already has some V4 Pro traffic in production.

  1. Inventory the callers. List every workflow, script and vendor integration that sends deepseek-v4-pro, with its monthly token volume by type. If the model string is hardcoded in more than one place, fix that first so the next migration is one edit.
  2. Freeze an evaluation set. Twenty to fifty real tasks per workflow with known good outputs. Include the knowledge-heavy and long-context cases, because those are where V4 Pro's base model still leads.
  3. Run V4.1 Flash at three effort levels. Low, mid and high, recording accuracy, output tokens and latency at each. Pick the lowest level that passes.
  4. Compare cost per passing task, not cost per token. A model that needs two attempts is not cheaper at half the price.
  5. Test the vision path if you have one. If a workflow currently runs OCR or a separate vision model before V4 Pro, try V4.1 Flash on the raw image; DocVQA at 95.6 on the model card suggests it may replace the step.
  6. Re-tune rate limiters and retries. The concurrency limit rises from 500 to 2,500. Client-side throttles set for V4 Pro will leave throughput unused.
  7. Schedule flexible work off-peak. Peak is 06:30 to 09:30 and 11:30 to 15:30 IST on weekdays. Anything batchable belongs outside those windows at half price.
  8. Cut over per workflow, not all at once. Move the workflows that passed, keep the rest on V4 Pro while it lasts, and log the model name on every request so the two are separable in analytics.
  9. Watch the two DeepSeek pages weekly. The retirement plan changed within 24 hours in September 2026. The next change may be as fast.
If V4 Pro routing does switch on in 2026

DeepSeek's 10 September plan was that deepseek-v4-pro requests would be served by V4.1 Flash at V4.1 Flash rates. If that plan is reinstated, callers who have not migrated will get the new model silently, with different vision support, a different effort dial and different output behaviour, at a lower price. Silent is the problem. A team that has already run the evaluation set knows what it will get; a team that has not will find out from its customers.

When Should a Team Stay on DeepSeek V4 Pro in 2026?

When its own evaluation set shows V4 Pro ahead on the workflow and the cost difference does not matter at that workflow's volume. On DeepSeek's base-model table V4 Pro leads on fact recall (SimpleQA-Verified by 13 points), multilingual knowledge (MultiLoKo by 5 points) and long-context understanding (LongBench-V2 by 6 points). A workflow that depends on the model knowing things rather than doing things, in a language other than English or Chinese, or across very long documents, is the candidate for staying. Everything else is a candidate for moving. Either way, staying is a decision with an expiry date, because DeepSeek has stated the destination.

What Are the Common Migration Mistakes in 2026?

Key Takeaways for 2026

Distk runs model migrations for growth teams across India and internationally as configuration changes gated by evaluation sets, so that the next vendor notice, however short, is an afternoon's work. If you have DeepSeek V4 Pro in production in 2026, that migration plan is where we start.

DeepSeek V4 Pro to V4.1 Flash in 2026: FAQs

Is DeepSeek V4 Pro being shut down on 14 September 2026?

The 10 September announcement said all V4 Pro requests would route to V4.1 Flash from 04:00 UTC on 14 September. The pricing page updated 11 September says V4 Pro service continues after that date with billing unchanged, pending further notice. Treat V4 Pro as end-of-life with no fixed date.

What model name replaces deepseek-v4-pro?

deepseek-flash, which serves DeepSeek V4.1 Flash. The legacy names deepseek-v4-flash and deepseek-v4-flash-vision-exp also resolve to V4.1 Flash at Flash prices.

What changes for my code when I move to V4.1 Flash?

Same endpoints and feature list. Differences: native image input, reasoning effort as an integer 1 to 100, thinking on by default, a 2,500 concurrency limit instead of 500, and Flash pricing. Self-hosters also get a new prompt format.

Will V4.1 Flash be worse than V4 Pro on my workload?

On DeepSeek's tables it is ahead on most agentic rows and behind on fact recall, multilingual knowledge and long-context understanding. Only your own evaluation set can answer it for your tasks.

How much will I save moving from V4 Pro to V4.1 Flash?

Per DeepSeek's rate card: about 7 times on cache-hit input, 4.4 times on cache-miss input and 3.3 times on output. Actual savings depend on your token mix and on the reasoning effort you set.

Should I migrate all traffic at once?

No. Cut over per workflow after each passes its evaluation set, log the model name on every request, and keep remaining workflows on V4 Pro while it is available.

Make the next vendor notice an afternoon's work

Distk turns model migrations into configuration changes gated by evaluation sets: inventory, test set, effort tuning, per-workflow cutover, logging. If DeepSeek V4 Pro is in your 2026 stack, we make sure the switch, whenever it comes, is one you already tested.

Start the conversation →