ReelShort vs FlexTV: The Industry Argument, Made Concrete

Updated August 19, 2026 · Tested by the AI Drama Reviews editorial team · 37 min read · ReelShort 9.5 / 10 · FlexTV 9.2 / 10

Which is better, ReelShort or FlexTV?

Every other comparison on this site matches two apps doing the same thing at different prices. This one does not. ReelShort and FlexTV sit on opposite ends of the single question the whole category is arguing about, which is whether a vertical drama needs to be filmed at all. ReelShort's answer is that it does, and its budget goes into a small English-language slate shot with US casts. FlexTV's answer is that it does not, and it said so publicly.

FlexTV vice-president Tang Tang told MIT Technology Review on May 15, 2026 that moving North American production to AI cut costs by 80 to 90 percent, compressed schedules from three or four months to under a month, and let teams of about ten people do work that previously needed full crews. That is a vendor describing its own results, and we label it as such throughout. It is also the most specific on-the-record account anybody in this industry has given.

What a viewer gets for that difference is the whole question, and the honest answer is that it depends almost entirely on genre. Fantasy, xianxia, werewolf and superpower stories survive AI production well. Contemporary human drama does not, and contemporary human drama is a great deal of what ReelShort makes well. We assess each app on its own in the ReelShort review and the FlexTV review; this page carries the choice between them.

ReelShort vs FlexTV at a glance

ReelShort is the expensive, narrow, natively English option; FlexTV is the cheap-to-produce, fast-refreshing, generated one. Everything in the table below follows from that one strategic split. ReelShort spends a large budget on a small slate and prices access high. FlexTV spends a small budget on a large slate and will not tell you what access costs.

One row in that table is doing more work than the others. ReelShort publishes something resembling a price list and FlexTV publishes nothing, and that difference is not cosmetic: it is the difference between being able to budget a season in advance and finding out at the paywall. It costs FlexTV 3.1 points on one of our six axes and it is the main reason a platform we find genuinely interesting sits below one we find expensive.

How each one is actually made

ReelShort commissions and films English-language originals with American casts. FlexTV generates its North American slate with AI, in teams of about ten people, in under a month per title. These are not variations on a production method. They are two different industries that happen to deliver into the same 9:16 player on the same phone.

The ReelShort model: a small slate, shot properly

ReelShort, operated by Crazy Maple Studio and launched in 2022 by the vendor's own account, is the only app in our thirteen whose flagship product is filmed in English by American performers for an American audience. That means a script, a cast, a crew, locations and a schedule, and it means the slate has to be small because each title carries a real budget. Some ReelShort titles carry a Dubbed label, which marks them as localised imports rather than originals; if the absence of a dub is your reason for choosing the app, check the label before you commit to a series.

The commercial result is visible in the reported figures. ReelShort recorded 49.9 million monthly active users in November 2025, roughly a 27% category share and about $600 million of North American revenue, with $130 million of Q1 2025 in-app revenue up 31% quarter on quarter (Sensor Tower). Sensor Tower's method estimates App Store and Google Play in-app purchases only and excludes advertising, third-party Android stores and direct web payments, so treat it as a floor rather than a total.

The FlexTV model: ten people and a month

FlexTV's pipeline, as described by its own vice-president, replaces the crew with a small team running a generation workflow. In practice that means story and series planning, a human-reviewed script and shot list, a locked character bible of reference images, shot generation in clips of roughly five to twelve seconds, synthetic voice and dubbing, then an edit pass that repairs artefacts and holds continuity together. Stage four is the part everyone pictures and it produces the least finished footage of any stage.

The reason the clips are short is a model constraint rather than a house style. ByteDance's Jimeng runs to about 12 seconds per generation, or 30 seconds with motion templates; Kling reaches around three minutes; Sora 2 and Runway Gen-4.5 top out at 1080p. A 90-second episode is therefore assembled from a dozen or more separate generations, and the working metric inside a studio like this is cost per accepted shot rather than cost per generation, because most of what comes out is discarded.

Two model releases sit behind the current quality level: Seedance 2.0 from ByteDance in February 2026, a notable step up, and Google's Veo 3.1, which adds reference images, first-and-last-frame control, scene extension and camera controls, and stamps a SynthID watermark on its output. Our full breakdown of the pipeline and the tools is in AI dramas vs live action.

What did AI production actually save?

Between 80 and 90 percent of production cost, on FlexTV's own account, and none of it has reached the viewer. Live-action vertical drama costs roughly $150,000 to $200,000 per screen hour. AI production runs $20,000 to $50,000 per screen hour. That is the entire economic argument of this page in two lines, and everything else is a consequence of it.

Converted to a per-minute figure, $150,000 to $200,000 an hour is about $2,500 to $3,300 per finished on-screen minute; that division is ours, not a reported number. At the other end, a heavily optimised Chinese pipeline has been described at around $30 a finished minute, which is not a typical US result and should not be read as one. Independent creator Jiang Lan accounted for roughly $280 a finished minute, or about $28,000 for twelve eight-minute episodes (TVBS, August 7, 2026). NOW News reported on December 15, 2025 that a complete 60 to 80 episode AI microdrama can be produced for $3,000 to $14,000.

What the headline percentage hides

An 80 to 90 percent saving on generation is not an 80 to 90 percent saving on making a show. The cost lines that survive are failed generations, retakes, upscaling, voice correction, human editing, music and sound effects, storage, translation and quality control, distribution fees, legal review, refund handling, and above all paid user acquisition, which is now often the largest single line item in this business. None of the per-minute figures on this page include it.

That matters for the comparison because it means the two apps' cost structures converge more than the headline suggests. ReelShort and FlexTV both have to buy their audience in the same auction, at the same prices, from the same platforms. The production saving is real and it is large; it is also a saving on the smaller half of the problem.

Did the saving reach you? No

FlexTV's production costs fell by up to 90 percent and its US retail price did not become public, let alone lower. ReelShort, which pays full freight for cast and crew, publishes a price list. There is no observable relationship in this category between what a title costs to make and what a viewer is asked to pay, because retail price is set by paywall design and willingness to pay, not by cost of goods.

We flag this because the cheaper-production-means-cheaper-viewing claim circulates widely and nothing supports it. If a reader's real question is money rather than technology, neither of the apps on this page is the answer, and we say where the answer is in the section on what neither one does. Full category arithmetic sits in our guide to what AI dramas actually cost.

What the difference looks like on screen

On ReelShort it looks like nothing, which is the point: performances land as written because the performers are speaking English on camera. On FlexTV it looks like short shots, faces that shift subtly between cuts, and hands you learn not to look at. Both platforms are competent at what they do. They are competent at different things.

ReelShort scores 10.0 on our production and dub quality axis and FlexTV scores 9.0, and that single point is doing a lot of work. It is not a claim that ReelShort's writing is better, because often it is not. It is a claim that nothing sits between the performance and you: no dub to drift against lip movement, no translated dialogue to read, no generated face to hold together across a cut.

The FlexTV artefact list, in the order you will notice it

Identity drift is first and worst: a face changes subtly between shots because each shot was generated separately. Then object permanence, where props appear and vanish between cuts, and clothing, jewellery, hairstyle or apparent age shifting without explanation. Then hands, which deform into extra or fused fingers under motion. Then background text, which resolves into pseudo-letters that look like language from a distance and nothing at all up close.

After that the subtler ones: lip movement out of sync with dialogue, screen direction reversing so two characters appear to swap sides of a conversation, physics that do not hold, and narrative continuity breaking across episodes. There is one more that no single shot reveals, which is that models trained on similar data produce similar faces, so unrelated series start to feel cast from the same handful of people.

What the mitigations do and do not fix

Character reference images, a locked character bible, fixed seeds, first-and-last-frame controls, shorter shots, shot-level human review and continuity editing all reduce these artefacts. None of them eliminates any of them. The academic literature is unambiguous on this point: spatiotemporal consistency in generated video is an open problem, surveyed by ACM in May 2026 and by OpenReview in March 2026, and a survey on the March 2026 state of long-video narrative consistency reads as a list of things that do not yet work.

The practical consequence for a viewer choosing between these two apps is simple. On FlexTV the mitigations are why the fantasy titles hold together and the contemporary ones do not, because a mitigation that shortens shots and avoids sustained close two-person dialogue is invisible in a xianxia duel and destructive in a kitchen-table argument. Our explainer on why AI drama characters change goes through the mechanism.

Which genres survive AI production?

Fantasy, xianxia, werewolf and superpower stories survive AI production well. Contemporary human drama does not, and contemporary human drama is much of what ReelShort makes well. That sentence is the most useful thing on this page, because it converts an abstract argument about technology into a decision you can make from your own taste in ten seconds.

The mechanism is not mysterious. Generated video fails hardest at exactly the things realism requires: a face holding identity through a long close-up, hands doing ordinary tasks, two people talking across a table for ninety seconds, a room that stays the same room. It fails least where the world is already unreal, where cuts are fast by convention, where costume is spectacular enough to carry the shot, and where an audience arrives expecting stylisation.

One reviewer's experience is worth setting against that table because it is the closest thing to a controlled test anyone has published. Jenny Cooper of Vertical Drama Love reported on June 24, 2026 that she watched 25 AI titles and finished three, concluding that story remains decisive and that fantasy was the strongest category. That is one reviewer's stated experience rather than survey data, and we cite it as such. It also matches what our own browsing found, which is why we repeat it.

Read the two columns as a rule rather than row by row: if the story could plausibly happen to someone you know, install ReelShort; if it could not, FlexTV is the better use of your evening and possibly your money. Our genre guide covers what each shelf actually contains across all thirteen apps.

What will each one cost you?

ReelShort publishes the only official pricing page in the category, and even that is a promotional help article with no exposed update date. FlexTV publishes no US price at all, which is why it scores 5.5 on billing transparency, one of the lowest marks anywhere on this site. One app tells you imperfectly. The other does not tell you.

ReelShort's page lists coin bundles at about $1.99 to $49.99, weekly VIP at roughly $14.99 to $19.99, monthly VIP at about $39.99 and an annual plan at about $199.99. Unlocking one full series with coins cost $30 to $50 in hands-on testing, which is roughly $0.50 per minute watched. That page was published and indexed as of August 2026; the absence of an update date is why the axis scores 8.6 rather than 10. Verify at checkout on both apps.

What a missing price actually costs you

An unpublished price is not a neutral fact and we refuse to score it as one. It means you cannot budget a season before you start one, cannot compare FlexTV against anything, and cannot tell whether the number at the paywall is normal or a regional experiment. The only defence available is to open the store page, read the in-app purchase list before installing, and treat whatever it says as the real price.

Across our full ranking of thirteen apps, nine publish no usable US figure at all, so FlexTV is typical rather than exceptional. ReelShort is the exception, and the exception is an undated marketing page. That is the state of the category in August 2026 and it is worth naming plainly rather than working around.

The renewal is more expensive than the choice

Both apps bill through the app stores and both cancel there rather than in the app. On iPhone: Settings, tap your name, Subscriptions, then the app, then Cancel Subscription. On Android: Play Store, profile icon, Payments and subscriptions, Subscriptions, then the app. Switching something off inside the app frequently does not stop the billing.

A ReelShort weekly plan left renewing at $19.99 is about $86 a month for a service most people intended to use once. Business Insider reported on January 28, 2026 that around 30% of surveyed ReelShort users pay $25 to $100 a month while about 27% pay nothing at all, which is roughly the distribution a weekly auto-renewal model produces. Cancel on the day you subscribe; you keep the week you paid for.

How did we reach 9.5 and 9.2?

The 0.3-point gap comes from two axes pulling in opposite directions. ReelShort wins production quality, worth 20%, by 10.0 to 9.0, and billing transparency, worth 10%, by 8.6 to 5.5. FlexTV wins cost to finish a season, worth 25%, by 9.0 to 7.6, and both catalogue and free tier. On the mechanical arithmetic these two apps are almost tied.

The arithmetic, so you can check it

Weighting those axes in the published proportions gives 8.78 for ReelShort and 8.72 for FlexTV, a mechanical gap of 0.06 points. Our published verdicts are 9.5 and 9.2, a gap of 0.3. Both sets of numbers are printed here rather than only the flattering one, and the ordering is the same either way, but the honest reading is that the mechanical calculation puts these two apps within a rounding error of each other.

Why the published gap is wider: production quality and billing transparency are the two axes where a reader's own experience diverges most from a weighted mean. A viewer who cannot stand generated faces loses everything on FlexTV regardless of price, and a viewer who cannot find out what an app charges is exposed in a way that a 10% weight understates. We widen the gap deliberately and we show you the number we widened it from.

One limitation applies harder on this page than on most. There is no standardised independent quality benchmark for AI-drama seasons, and no dataset shows what percentage of any app's catalogue is AI-generated. The production-quality and catalogue axes are structured judgements from watching both apps in August 2026, not measurements. The billing-transparency axis is the only one here that is purely a matter of fact, and it is the one FlexTV fails.

Catalogue size and refresh rate

FlexTV, on refresh rate, by the same mechanism that produced its cost saving: a title that takes under a month to make can be replaced monthly, and a filmed original cannot. ReelShort's slate is small by design because English-language originals with US casts are expensive, and the same budget cannot buy both depth and breadth.

Neither company publishes an audited catalogue count and no third party measures one, so we are describing browsing depth rather than quoting a number. FlexTV scores 9.2 on catalogue and refresh against ReelShort's 8.8, and that 0.4 is among the softer numbers on this page.

What the refresh advantage buys is contested. In a category where DataEye counted about 1,200 new AI titles a day on Douyin in the first half of 2026, more shelf is not obviously better shelf, and a narrow commissioned slate does filtering work that a fast-refreshing one pushes back onto you. If you value a high hit rate on the first title you tap, ReelShort's smallness is a feature.

The catalogues do not overlap and there is no cross-app search, which is true of every pair of apps in this category. Every platform finances its own titles and none licenses to another, so the real choice is which library you are not watching this month rather than which version of a library you prefer. Full app-by-app comparison on our reviews hub.

Whose free tier goes further?

ReelShort's is the one we can describe: the first 5 to 10 episodes of most series, then a wall, with no daily refill and no ad-unlock route back in. FlexTV reports free episodes and publishes no allowance, so we cannot tell you what you get until you open it. FlexTV scores 9.2 on this axis and ReelShort 8.9, and the gap is about what the free tier is for rather than about episode counts we cannot compare.

The category convention is the first three to ten episodes before a coin wall, a subscription prompt or an ad-unlock appears. Both apps place that wall at the same moment, and it is not a natural break: episodes are built with a hook in the first three seconds and a cliffhanger at the end precisely so the wall has somewhere effective to sit. Knowing it is coming is most of the defence against it.

Use whichever free allowance you get to answer one question rather than to watch one story: does generated video bother you? Spread the episodes across three different FlexTV series, include at least one contemporary-set title alongside the fantasy, then watch thirty seconds of a ReelShort original for comparison. An hour of that outranks our scores, and it costs nothing.

Did cheap production make better shows?

No. It made failures cheap, which is a different achievement and a smaller one. DataEye counted roughly 221,900 AI dramas published on Douyin in the first half of 2026, about 1,200 a day, and found that only around 1,055 of them, or 0.47 percent, passed 100 million views. Output rose by orders of magnitude and the hit rate did not rise with it.

The trend behind that figure is steep. In January 2026 Douyin was seeing about 470 AI titles a day; by April 2026 the platform carried roughly 44,200 AI titles against 3,248 live-action ones in a single month. The China Netcasting Services Association reported that more than 95% of new Chinese microdramas in Q1 2026 were AI-generated. This is Chinese platform data, the only market where AI-drama output is measured at scale, and US catalogues are stocked substantially from that pipeline.

What that means for a reader choosing between these two apps is a warning against reading FlexTV's cost advantage as a quality advantage. Cheap production removes the financial penalty for making something nobody watches, which is why so much of it now exists. It does not improve the odds that any individual title is good, and the 0.47% figure is the closest thing to a measurement of those odds that anyone has published.

Set against that, one honest counterpoint. Bogdan Nesvit of Holywater told TheWrap in 2026 that engagement metrics for AI content were in some cases comparable to live-action, and a former Meta executive described in April 2026 what is emerging as not a substitute, but a parallel category, one defined by speed, variation, and the ability to cater to almost any audience preference imaginable. A category with a 0.47% hit rate and enormous volume can still be a real category. It is just not the same category ReelShort is in.

What happened to the people who used to make these?

Reporting from the Chinese market describes microdrama pay falling by half or more, with some studios shrinking by around 70 percent. FlexTV's own figure of teams of about ten people replacing full crews is the same fact told from the employer's side. This section exists because a comparison page that discusses an 80 to 90 percent cost reduction without saying where the money used to go is not an honest page.

We will say it plainly: a reader who objects to FlexTV on labour grounds has a coherent position, and this page will not try to talk them out of it. The saving is not an efficiency conjured from nowhere. It is largely wages that are no longer paid, to crew, to background performers, to the tier of working actors for whom microdrama was a reliable job. If that is your reason for installing ReelShort instead, it is a better reason than any of our axis scores.

The US protections that exist do not reach this. SAG-AFTRA's framework defines a digital replica as a program using a performer's voice, image or performance to generate new performances and requires consent, disclosure, compensation and control, with materials dated November 17, 2025 and interactive-media protections covering digital voice replicas from July 2025. The 2023 WGA agreement limits requiring writers to use AI and addresses the status of AI-written material. Both are contracts between US unions and signatory employers, and a generation pipeline staffed by ten people outside that jurisdiction is not a signatory to either.

State law reaches slightly further and still not far enough for this case. Tennessee's ELVIS Act, in force July 1, 2024, covers name, image, voice and likeness plus the technologies used for unauthorised replicas. California's AB 1836 and AB 2602 address digital replicas including of deceased performers, and contracts. The FTC's impersonation rule covers government and business impersonation from April 1, 2024, with the extension to private individuals not finalised. None of that governs a synthetic performer who resembles nobody in particular.

The counter-argument deserves stating too, because it is not empty. Independent creators who could never have afforded a crew can now finish a series for $3,000 to $14,000, and revenue splits are commonly 50/50, with a title reaching a billion views returning $100,000 to $500,000 to its creator. That is a real redistribution toward people who previously had no access. It is also not the same people who lost the wages. Our full treatment is in will AI replace actors.

Who should pick ReelShort?

Install ReelShort if you want drama about people, if performance is the reason you watch at all, or if you have already tried generated video and it broke the spell. That is most people who ask us this question, which is why ReelShort scores higher despite costing more and telling you less about its catalogue than FlexTV tells you about its production.

• You watch for the acting. English-language originals with American casts mean no dub to drift, no translated dialogue to read, and no generated face to hold together across a cut.
• Your genres are contemporary. CEO romance, revenge, family conflict, anything set in an ordinary American house. These are precisely the genres AI production damages most.
• You want a high hit rate. A small commissioned slate does filtering that a fast-refreshing generated catalogue cannot.
• You want to know the price before you pay it. ReelShort's page is undated and promotional, and it is still the only one in the category.
• You object to AI production on labour grounds. ReelShort is the least AI-produced catalogue among the major apps we tested, and that is a legitimate basis for choosing it.
• You watch one series at a time and finish it. A single monthly VIP at about $39.99 is a defensible budget for that pattern; the coin route is not.

Skip ReelShort if fantasy and xianxia are what you came for, if you watch in volume, or if $30 to $50 for one series sounds alarming rather than normal. The full ReelShort review carries the axis-by-axis detail and the honest list of what it does badly.

Who should pick FlexTV?

Install FlexTV if you want to see what an AI-produced slate actually looks like at scale, or if your genres are fantasy, xianxia, werewolf and superpower stories, which is where generated video is at its most convincing. FlexTV is the clearest example in the category of the thing everyone is arguing about, and no amount of reading substitutes for twenty minutes of watching it.

• You want the argument settled by your own eyes. FlexTV is the platform that said publicly what it did. Nothing else on the market is this legible.
• Your genres are unreal. Cultivation fantasy, mythology, superpower systems and shifter romance all survive generation, and some of them are better served here than anywhere.
• You want volume and fast refresh. Sub-one-month production schedules put new titles on the shelf at a rate a filmed slate cannot match.
• Artefacts do not bother you. If short shots and slightly unstable faces read as a style rather than a defect, the price of admission drops to nothing.
• You are researching the industry. For anyone writing about, investing in or working in this business, FlexTV is the primary source and ReelShort is the control group.

Skip FlexTV if you need a price before you commit, if you watch for performances, if contemporary drama is your shelf, or if the labour argument above settles it for you. The full FlexTV review sets out why billing transparency scores 5.5 and what that absence costs you in practice.

What neither one does

The failings ReelShort and FlexTV share are larger than the argument between them, and a page that only crowns a winner is not being useful. Six problems apply to both, and to most of the eleven other apps we rank.

Neither labels AI content

Neither app tells you which titles were generated, which were dubbed by a voice model, or what share of the catalogue is either. There is no filter and no per-title marker beyond ReelShort's Dubbed label. Viewers who want AI-produced drama cannot reliably find it and viewers who want to avoid it cannot reliably avoid it, and both groups demonstrably exist. Our detection guide exists because the apps will not do this for you.

Neither publishes a price you can hold them to

FlexTV publishes nothing. ReelShort publishes a promotional help article with no exposed update date. Nine of our thirteen ranked apps publish no usable US figure at all. Neither of these two is an exception worth praising.

Both put the paywall where it hurts most

Both place the wall mid-cliffhanger, at the point where willingness to pay peaks, which is the business model of the category rather than an accident of scheduling. Episode structure exists to serve it: hook in three seconds, rapid conflict, frequent reveal, cliffhanger ending.

Both cancel in the store, not in the app

Neither app can reassure you inside itself that the billing has stopped, because cancellation lives in Apple's or Google's subscription screens. Weekly plans renew fast and quietly, and this is the most expensive design decision in the category.

Both are poor on a television

Both are 9:16 vertical video, correct for the format and wrong for a horizontal screen, which fills with black bars either side of a narrow strip. Both run on Android, iOS and a desktop browser. ShortMax is the only app in our ranking with genuine TV coverage and it is worse than both of these on other axes.

Neither is the cheap option, and we are paid by the app that is

If your real question was money rather than technology, the answer is neither of these. DramaBox runs from about $5.99 a week for unlimited viewing against ReelShort's $30 to $50 in coins for one series, with the largest catalogue in the category and a free tier that refills daily. We earn an affiliate commission from DramaBox and nothing from either app on this page, so treat that recommendation with the scepticism it deserves and note that DramaBox gives you a mixed catalogue rather than the AI-first one FlexTV offers. The whole set of head-to-head pages is at our comparisons hub.

Frequently asked questions

The bottom line

ReelShort takes it at 9.5 to 9.2, and the mechanical arithmetic behind those numbers is 8.78 to 8.72, which is a tie. ReelShort wins because filmed performance is what most people actually want from drama, because it is the only app in the category that publishes a price, and because contemporary human stories are the genre AI production damages most and the genre ReelShort is built on. It costs about $19.99 a week, $39.99 a month, $199.99 a year, or $30 to $50 in coins for one series, and that is a lot.

FlexTV is not the loser of this comparison so much as the answer to a different question. It moved its North American production to AI, cut costs 80 to 90 percent by its own account, compressed schedules from three or four months to under one, and runs teams of about ten people. If your genres are fantasy, xianxia, werewolf or superpower stories, that trade costs you very little and buys you a faster-refreshing shelf. If your genres are contemporary, it costs you the thing you came for.

Two facts should travel with any decision made from this page. The first is that cheap production did not make hits more common: DataEye counted about 221,900 AI dramas on Douyin in the first half of 2026 and roughly 0.47% of them passed 100 million views. The second is that the 80 to 90 percent saving was largely wages, and a reader who declines to fund that has a coherent position we are not going to argue with.

We earn nothing from either app here. That is unusual enough on a comparison page to be worth repeating at the end: nothing on this page is steering you, and the only recommendation we are paid for is the one we flagged as such. Watch an hour of each free tier, ask yourself at the first paywall whether you are annoyed or relieved, and trust that over any score on this site.

DramaBoxEditor's pick — biggest short-drama library, lowest price to go unlimited 9.8Excellent

• Thousands of series, roughly 200 new titles every month
• Free episodes daily, plus ad-unlocks — no card needed to start
• Unlimited plans from about $5.99 a week, the lowest entry price we found
• Runs on Android, iOS and in a desktop browser

Go to DramaBoxAffiliate link · free to download · in-app purchases available

Affiliate link. We may earn a commission if you install through it, at no extra cost to you — it never changes our scores or placements. How we test · Disclaimer

Related reading

Sources

• MIT Technology Review (May 15, 2026) — Tang Tang, vice-president of FlexTV, on the move of North American production to AI: costs down 80 to 90 percent, schedules from three or four months to under one month, teams of about ten people replacing full crews. This is a vendor describing its own results and is not independently audited.
• ReelShort — official site and its vendor pricing help page, published and indexed as of August 2026 with no update date exposed: coin bundles ~$1.99–$49.99, weekly VIP ~$14.99–$19.99, monthly VIP ~$39.99, annual ~$199.99. Roughly $0.50 per minute watched via coins; $30–$50 to unlock one full series in our own testing.
• FlexTV — official site and in-app store listings, browsed August 19, 2026. No official US price page exists and no US price could be confirmed. We treat that absence as a finding and score it 5.5 rather than estimating a number.
• DataEye (2026), including via MIT Technology Review (May 15, 2026) — roughly 221,900 AI dramas published on Douyin in H1 2026, about 1,200 a day; roughly 1,055 of them (0.47%) passed 100 million views; about 470 AI titles a day in January 2026; about 44,200 AI titles against 3,248 live-action in April 2026.
• China Netcasting Services Association via Global Times (2026) — more than 95% of new Chinese microdramas in Q1 2026 were AI-generated. Chinese market data; the US has no equivalent measurement.
• Sensor Tower, reproduced in State of Short Drama Apps 2025 (November 25, 2025) — ReelShort Q1 2025 in-app revenue of $130 million, up 31% quarter on quarter. Methodology: App Store and Google Play in-app purchases only, excluding advertising, third-party Android stores and direct web payments.
• Sensor Tower via Media Play News (April 27, 2026) and eMarketer (March 5, 2026) — ReelShort US Q4 2025 engagement of 35.7 minutes per user per day, against Prime Video 26.9, Netflix 24.8, Disney+ 23. An intensity measure per app user, not total viewing time.
• ReelShort operator Crazy Maple Studio, vendor-stated — 49.9 million monthly active users in November 2025, roughly 27% category share, about $600 million of North American revenue, service launched 2022.
• Business Insider (January 28, 2026) — survey reporting that about 30% of ReelShort users questioned pay $25 to $100 a month while about 27% pay nothing. Cited as survey data; sample and method not restated here.
• NOW News (December 15, 2025) — an AI microdrama of 60 to 80 episodes produced for $3,000 to $14,000. TVBS (August 7, 2026) — independent creator Jiang Lan at about $280 per finished minute, roughly $28,000 for twelve eight-minute episodes.
• Vertical Drama Love, reviewer Jenny Cooper (June 24, 2026) — 25 AI titles watched, three finished, fantasy the strongest category, story decisive. One reviewer's stated experience, not survey data.
• PetaPixel and Today (August 2026) — director Qingge Gao on viewer indifference to AI production. TheWrap (2026) — Bogdan Nesvit of Holywater on engagement metrics for AI content. Both cited as expert opinion, not measurement.
• ACM, A Survey: Spatiotemporal Consistency in Video Generation (May 18, 2026); OpenReview on narrative and temporal consistency in long video (March 26, 2026); arXiv:2510.04999, Bridging Text and Video Generation: A Survey (October 6, 2025) — the technical basis for the artefact list on this page.
• SAG-AFTRA digital replica materials (November 17, 2025) and July 2025 interactive-media protections; WGA 2023 MBA; Tennessee ELVIS Act (in force July 1, 2024); California AB 1836 and AB 2602 (2024); FTC impersonation rule (in force April 1, 2024, private-individual extension not finalised) — the labour and likeness framework cited in the labour section.
• Our own testing — both apps installed on US iOS and Android accounts, free tiers exhausted, ReelShort paid through with our own money and the receipt recorded, cancellation completed in the store, August 2026. Protocol on how we test.

Declared gaps. No authoritative dataset shows what percentage of either app's catalogue is AI-generated; our classifications are assessments, not measurements. No US benchmark exists for the cost of a completed, edited, legally cleared AI-drama episode, so every cost figure here describes production elsewhere or a vendor's own accounting. No standardised independent quality benchmark ranks AI-drama seasons. No representative US complaint dataset exists, only recurring qualitative themes. FlexTV publishes no US price at all. We state these rather than filling them.

Affiliate disclosure. We have no affiliate relationship with ReelShort and none with FlexTV, and we earn nothing whichever of them you install or if you install neither. Our only paid relationship in this category is with DramaBox, which is not one of the apps compared here; it is named once, in the section on what neither app does, because it is the honest answer for a reader whose real question was cost, and that mention is flagged as commercial in the body of the page rather than only in this footnote. Commissions do not change our scores, our weights or our ranking. Accuracy notice. Prices, coin values, free-episode allowances and catalogues change without notice and vary by region, device, store and promotion. Everything here was verified on August 19, 2026. ReelShort's pricing page is a promotional help article with no exposed update date and FlexTV publishes no official US price page at all, so the price quoted at checkout is the only binding one. Vendor-stated production figures are labelled as such throughout. Where we could not confirm a figure, we say so rather than estimating. Independence. We are not affiliated with ReelShort or FlexTV and this is not an official support channel for either; for account or billing problems, contact the app directly.