Get started with Valmera
Upload a recording, describe the edit in plain English, and Valmera’s AI agent does it: it reads the transcript, watches the frames, cuts, captions and reframes, checks its own preview, then reports back. You revise by chatting and export a full-resolution MP4 from your original file. Creating an account and uploading are free; editing requires a subscription (see plans).
Before you start
- A video file: MP4, MOV, M4V, MKV or WebM, up to 14 GB and 3 hours. Both limits apply.
- An account: free. Editing requires a subscription.
- A browser: any modern browser on desktop or phone. Nothing to install.
Five-minute quickstart
- Create your account
Go to valmera.io/register and sign up with Google or your email address (you’ll get a sign-in code). Creating an account and uploading are free. Editing requires a subscription — see plans when you’re ready.
- Upload your footage
Open Studio. Keep the switch on LONG VIDEO to edit one video, or choose SHORTS to cut a long recording into several shorts. Drop your MP4, MOV, M4V, MKV or WebM file onto the page (up to 14 GB and 3 hours). The file goes straight to secure storage; keep the tab open until the upload finishes. Formats and limits.
- Let Valmera analyze it
Valmera builds a word-level transcript with speaker labels, finds shot changes and silences, reads any burned-in text, and makes a fast preview copy. Progress shows in Studio. You don’t have to wait: type your request now and the chat box confirms “I’ll start when analysis finishes.”
- Describe your first edit
Type what you want in plain English, or tap a suggestion such as “Cut the silences and remove the ums.” Say who the video is for, the format you need and anything that must stay.
- Review the preview
The agent edits, renders a preview, looks at its own frames and listens to the audio, fixes what it finds, then tells you what it changed. Watch the whole preview: listen at the cuts, check names in the captions and check the framing.
- Revise
Reply in plain English, for example “Restore the sentence about pricing and make the captions smaller,” or adjust the timeline and transcript by hand. Every change is a new version and your original file is never modified, so you can step back with Previous edit at any time.
- Export
Click Export MP4 at the top of the workspace, or Export full-quality MP4 under the agent’s reply. Valmera renders the selected version from your original file at full resolution; when it’s ready, the button changes to Download MP4. Export guide.
Analysis and rendering time depends on the length of your video; everything else above takes a few minutes of your time.
Example first prompts
A good first request names the job, the format and what must stay. Pick the one closest to your video, paste it, and adjust the details.
- Talking head or vlogCut the silences and remove the ums. Add captions. Keep the original framing.
- Short-form clipTighten this into a 60-second clip about the main point, add bold captions, and make it 9:16 for TikTok.
- Podcast or interviewRemove dead air and filler words, keep both speakers’ answers complete, and add podcast-style captions.
- Tutorial or screen recordingRemove the long pauses, zoom in when I click the Settings button, and make the cursor bigger.
- YouTube videoCut the silences, put my uploaded music quietly under the whole video, and add a title card at the start that says “How we ship faster.”
Type these in Studio chat, or ask Claude, ChatGPT, Cursor or another assistant connected to Valmera’s MCP server. More in the prompt library.
What to expect at each stage
| Stage | What you see | What you can do |
|---|---|---|
| Uploading | Upload progress for your file | Keep the tab open. For large files, editing can start from a lightweight preview copy while the original finishes. |
| Analyzing | Analysis progress; the transcript appears when ready | Type your request. It’s sent automatically when analysis finishes. |
| Editing | The agent’s progress in the chat | Wait for the reply. Manual timeline changes pause while the agent works. |
| Preview ready | A new preview in the player and a reply listing what changed | Watch it, then reply with changes or export. |
| Exporting | Export MP4 shows Exporting with a percentage | The render runs on Valmera’s servers; come back to Download MP4 if you leave. |
| Done | Download MP4 | Download the file and post it anywhere. |
Revise the edit
Reply with one or two specific changes at a time. The agent edits the current version; it doesn’t start over, and it keeps everything you didn’t mention.
- The cut before “here’s the example” is too tight. Give it a little more room.
- Spell the product name “Valmera” in the captions.
- Restore the part about pricing and keep everything else.
- Make the captions smaller and keep them clear of my face.
- Go back to how the intro was before the last change.
Prefer to do it yourself? Drag the edges of a block on the timeline to trim it, use the pencil on a transcript line to fix a word, and use Previous edit and Next edit next to the player to compare versions. See timeline, transcript and versions.
Export your video
- The export renders the version you selected from your original upload at full resolution: an H.264 MP4 with AAC audio.
- Previews use a lower-resolution working copy for speed, so the final file is sharper than the preview.
- Exports on a paid plan have no watermark over your footage and end with a brief Valmera end card.
- Captions are burned into the picture. There’s no separate SRT or VTT file.
- Valmera doesn’t post to YouTube or TikTok for you: download the MP4 and upload it wherever you publish.
Details and troubleshooting: Export.
Prefer Claude, ChatGPT or Cursor? Use MCP
Valmera’s MCP server lets your AI assistant edit with the same tools Valmera’s agent uses, including the final export.
- Add Valmera to your assistant with the endpoint
https://valmera.io/mcp/server. Exact steps for each app are in MCP setup. - Sign in with your Valmera account (email or Google) and approve access.
- Ask: “Open my latest Valmera project, cut the silences, add captions, and show me the preview.”
- When you’re happy: “Export the final MP4 and give me the download link.”
Your assistant edits with the tools directly; it can’t hand the job to Valmera’s in-house agent. Edit a project in one place at a time: Valmera refuses a second editor while Studio or an assistant is working on it.
Other ways to start
Turn a long video into shorts
Before uploading, switch Studio to SHORTS, then upload a long recording such as a podcast or webinar. Optionally add a style reference (a short you like) and press Make shorts. Valmera picks strong self-contained moments, reframes them to 9:16, captions them and renders each as its own project you can open and edit. Export them one by one, or all at once with Export ready shorts. If the source is under a minute, Studio offers Edit as one short instead.
Build from photos and clips
Upload images or clips and ask for a slideshow or montage: “Make a 30-second slideshow from these photos with slow zooms and my music.” See slideshows from photos.
Start from a link
Paste a YouTube, Vimeo, TikTok, Dropbox or direct file link into the chat and say how to use it. Link imports have their own limits; see link imports.
Troubleshooting your first edit
| Symptom | Fix |
|---|---|
| “Upgrade to use the Valmera AI editor” after you send a request | Editing requires a subscription. Your request stays in the chat box; choose a plan, then send it. |
| The chat box says it will start when analysis finishes | Analysis is still running. Your request is queued and starts automatically. |
| The agent removed something you wanted | Ask “Restore the part where I say ‘…’.” Every cut can be restored without undoing other changes. |
| A name is misspelled in the captions | Ask “Spell it ‘Valmera’ in the captions,” or fix the sentence with the pencil in the transcript panel. |
| The preview looks soft | Previews use a lower-resolution working copy. The export renders from your original at full resolution. |
| Export shows “Finishing edit…” | The agent is still working on that version. Wait for its reply, then export. |
| “You’re out of credits” | Subscriber credits refill daily and monthly. See Credits for how the pools work. |
| “Message too long” | Keep a single request under 4,000 characters. Split a long brief into two messages. |
Limits
- Valmera edits footage you provide. It isn’t a text-to-video generator.
- Captions are burned in; there’s no SRT or VTT import or export.
- No team seats or shared editing yet.
- No direct publishing to YouTube or TikTok. Download the MP4 and post it.
- No custom font uploads, true crossfade dissolves or motion-tracked stickers.
- No audio denoise (“studio sound”) and no AI music generation.
- English is the best-tested transcription language.
Frequently asked questions
How long does analysis take?
It depends on the length of the recording: a short clip is quick and a multi-hour file takes longer. You don't have to wait to start. Type your request while analysis runs and the agent begins as soon as it finishes.
Do I need video editing experience?
No. You describe the result in plain English and the agent does the timeline work. A timeline, transcript and version history are there when you want direct control.
Will Valmera change my original video file?
No. Every edit is a new version of a non-destructive edit decision list (EDL). The uploaded file is never modified, any cut can be restored, and the export is rendered from the original.
Is Valmera free to use?
Creating an account and uploading are free. Editing requires a subscription; current plans are on valmera.io/subscribe.
Can I use Valmera on my phone?
Yes. Studio runs in mobile browsers, including uploading, chatting with the agent, previewing and exporting. A larger screen is easier for detailed timeline work.
What languages does Valmera understand?
English is the best-tested transcription language. For other languages, check names and captions closely and correct any line in the transcript panel.
Can ChatGPT or Claude edit my video with Valmera?
Yes. Connect Valmera's MCP server at https://valmera.io/mcp/server and your assistant can open your project, edit it with Valmera's tools, render a preview and export the final MP4.