InVideo AI Reels and TikTok: Turn Long Footage Into Vertical Clips

Yes, InVideo AI turns out Reels, Shorts and TikToks, and in November 2026 the strongest route starts from footage you already have. You upload a talk, a podcast or a product demo, and Agent Two reads a transcript, picks the hooks, cuts the pauses and builds each vertical clip on an editable timeline in the invideo Editor ^[1]. Every clip leaves as its own 1080p MP4. Timeline edits and exports cost no credits, while the transcript and any generated footage do. That split decides the bill. Clipping your own recordings stays light on any plan; Reels built from a prompt with premium video models want Plus or above. The agent never posts for you.

Cutting vertical clips from footage you already have

This is the workflow invideo documents in the most detail, and it suits anyone sitting on long recordings: webinars, interviews, podcasts, event talks. You hand Agent Two the raw file and the target, and it works like an assistant editor. It reads what was said, picks the hook and the key lines, removes pauses and filler words, and lays each clip on the timeline as ordinary clips you can retrim, replace or undo.

1. Upload the recording. The podcast workflow accepts MP4 or MOV video, or WAV or MP3 audio.
2. Tell the agent the clip is a Reel, a Short or a TikTok; invideo says it then reframes the footage for that platform.
3. Decide on approvals. Agent Two's default is Always allow generation; switch it to ask before any generation if you want to approve the transcript first.
4. Review the cut. Each clip sits on its own editable timeline, and a wrong opening line can be swapped by hand.
5. Export one 1080p MP4 per clip, at no credit cost.

The same pattern stretches to other material. In the podcast guide the full episode stays at 16:9 while the clips go out at 9:16 or 1:1 with captions. The sports highlight guide finds the big moments from crowd, whistle and commentator sound, can replay the key play in slow motion, and charges credits for a commentator transcript and paid stock music, while some tool ops can carry a charge too. A transcript is charged once per file, so pulling several clips from one upload does not pay for the words again.

Getting the frame right: 9:16, 4:5 and 1:1

Most source footage is wide, so the real work is reframing it, and the two routes cost very differently. The manual route is free: you set the clip to fill the frame, then adjust position and zoom until the speaker sits where you want. The generative route uses reframe models such as LTX 2.3 Pro or Luma Ray 3.2, which paint the picture missing outside the original frame, and invideo prices it by the length of the source clip.

Neither option tracks a face. The docs say generative reframing makes new picture rather than following a person around the frame. When the speaker stands near the center of a wide shot, a manual crop costs nothing and loses nothing important. Pay for generative reframing only when a straight crop would cut away something the viewer needs, such as a second guest at the edge or a product on the table. The cheapest fix happens at the shoot: keep your subject centered.

Captions, hooks and the safe zone

Captions come in two kinds. The default is a word-by-word caption track built from the transcript, where every word carries a start, an end and a speaker. It stays editable on the timeline until export, and you can switch it off for a single export. The alternative is a burned-in subtitle model, invideo Subtitles or VEED Subtitles, with styles that include TikTok and Instagram safe zones. Those are priced by video length and write the text into a new video, so you cannot fix a typo on the timeline afterwards.

Each app puts its buttons in a different place over a vertical feed, and captions under them get hidden. The Editor lets you add safe-zone guide lines that show while you work and do not render in the exported file; set them once per project before you place any text. For the opening seconds, invideo's blog advises a reel script of 60 to 90 seconds, with the hook inside the first three seconds and the problem stated in the next ten to fifteen. Treat that as the vendor's rule of thumb, not a guarantee of reach.

No footage? Building a Reel from a prompt

Agent Two, the default since August 2026, can build a Reel from nothing but a description, either as a conversation through the Script to Video or Explainer workflows or in Autopilot for a one-shot result. Either way the output is an editable timeline, not a locked file. invideo's Reel maker page describes a prompt box where you choose a creative strategy, a length and a media plan, and it advertises 100+ hook effects such as Drip Flip and Hero Walk; that count is invideo's own figure. Our InVideo AI text to video guide walks through the prompt route in full.

The original InVideo AI flow still works through the Older versions button on the prompt page. There you paste a script with Use my script, get a stock-media video with AI voiceover and subtitles, then edit it with Magic Box text commands; one of them changes the aspect ratio, and the video is generated again at 9:16. An SRT file can be downloaded after export.

For shop videos, invideo's TikTok Shop guide resizes product photos to 9:16 and animates them with Seedance 2.5 or Kling image-to-video models, but still tells you to lead with real footage, because a slideshow of zoomed stills is what TikTok Shop marks as low quality. UGC-style Reels can use real people or AI actors. InVideo AI requires explicit consent from the person in the recording, captured in a separate in-product step that can be withdrawn; celebrities and politicians cannot be cloned, and an avatar of anyone under 18 is forbidden even with parental consent.

What a Reel costs in credits, and which plan fits

Cutting, captioning by hand, reframing by hand and exporting are free; transcripts, generated clips, generative reframes, burned-in subtitles and paid stock spend credits. invideo says exports from the invideo Editor are free and carry no watermark on every plan, Free included, and its TikTok page puts that free export at 1080x1920 ^[2]. The Free plan's weekly limit on AI actions has no published size, so it works for trying a few clips, not for a posting schedule.

Starter costs $20 a seat a month billed monthly, with 400 credits a seat a month; it runs Agent Two Lite and generates video with Seedance 2.0 Fast and 2.0 Mini, but not Seedance 2.5 ^[3]. Plus costs $60 a seat a month billed monthly, with 2,000 credits and Seedance 2.5 at 1080p. At about 96 credits per 5-second Seedance 2.5 clip at 1080p, by our arithmetic Plus covers about 20 such clips a month and Max at 5,000 credits about 52, before retries. Retries cost again; invideo's own documented productions needed about 3 generations per usable shot.

So the plan follows the source. If your Reels come from recordings you already own, the paid parts are transcripts and the odd stock track, and Starter carries that load; Free is enough to test the Editor workflow first. If every Reel needs fresh Seedance 2.5 shots at 1080p, Plus is the lowest plan that can make them. Teams on shared workspaces can set a per-member credit limit, which stops one eager editor from draining everyone's month.

The annual card showed Starter at $10 a seat a month billed annually, as an offer shown in November 2026, and the year is charged in full upfront. Plan credits reset monthly on your subscription date and do not roll over. Fees are non-refundable; invideo's Terms allow a refund request within 24 hours of a first purchase or renewal only if no credits or premium features were used, and credits, including extra credits, are never refunded. Our InVideo AI pricing guide covers the annual offer and the full terms.

One edit, every platform, and the posting stays with you

The docs build platform versions as branches of the same timeline. The 9:16 Reel, the 4:5 feed cut and a 1:1 version each live on their own branch, the agent never changes the master timeline, and a fix to one shape cannot break another. Each branch exports one 1080p MP4. Branches do not merge, though, so a caption typo or a weak opening line has to be fixed on every branch that already carries it. Fix the master first, then branch.

When the files are ready, posting is your job. invideo's agents never post, publish, schedule or upload to social platforms or shops; its Shorts and TikTok Shop guides end with a file you download, upload yourself and check for any required AI disclosure. You can open the same project in the iOS and Android apps, but invideo says it has no app built for phone editing and concedes CapCut is better there. For the full timeline tour, see our invideo Editor guide; for long YouTube edits and podcast clips together, read InVideo AI for content creators.

The bottom line

InVideo AI earns its place in a short-form workflow when you already record long material. The bill grows with what you generate, not with how many clips you cut. Starter fits clipping your own footage, while Reels built with Seedance 2.5 start at Plus. Keep your subject centered, use the editable caption track, fix the master before you branch, and plan on downloading and posting every file yourself.