Last updated: 2026-11-05
Solmi is an AI music video generator from Singapore's VEGA GROUP that turns an uploaded MP3, WAV or Suno download into a beat-synced video in 3 to 10 minutes, by the vendor's count. It detects lyrics in 14 languages, can cast and lip-sync a singer from your photo, and runs on web, iOS and Android.
About Solmi
Solmi, run by VEGA GROUP PTE. LTD. in Singapore, makes music videos from a song you already have: upload an MP3, WAV or M4A of up to 200 MB, or the file you downloaded from Suno, and Solmi detects the tempo, key and lyrics, plans the shots and renders a finished video for YouTube, TikTok, Reels or Shorts. Shots render in parallel, so a full song does not wait on each scene in turn.
What sets it apart from Freebeat and Neural Frames is the singer. Add a clear, front-facing photo and Solmi casts that person as the lead and lip-syncs them to the vocal, with up to three reference photos for a duet or group. Each video is saved as a Studio project, so you can rewrite one shot's description or regenerate a single shot and keep the rest. Lyric detection covers 14 languages, and lyric captions can be switched on or off.
The same account carries a wider kit: a singing video tool that animates one portrait (the music cousin of a talking-photo tool like HeyGen), a dance video generator, a lyric video maker, an AI song generator, a vocal remover and stem splitter, plus image and video tools priced per generation. It also ships an MCP server, so a song or music video can be made from Claude, ChatGPT or Cursor, and a REST API for scripts.
It suits independent artists and Suno or Udio users who need a video to release a track. If you want visuals that react to individual stems, Neural Frames or Kaiber fit better. For shot-by-shot clip work, Runway or Pika give more control, and CapCut is the usual place to finish the edit. Our AI music video generators guide prices these side by side, and the AI music video directory lists the rest of the field.
Pricing is credit-based, with a free tier and three monthly plans. Read the terms before uploading an unreleased master: by default your files can feed Solmi's own model training, and switching that off is a setting you have to change yourself.
Pricing
Free plan: 100 one-time music video credits, enough for one 20-second 480p preview with a watermark, plus song credits that refill to 10 a day. Plus is $35 a month for 2,000 credits, which Solmi puts at roughly 3 to 7 one-minute music videos. Creator is $85 a month for 5,500 credits and Pro is $135 a month for 9,000 credits.
All paid plans render at 720p and 1080p, drop the watermark and include commercial use. One-time credit packs run from $30 for 1,500 credits to $600 for 35,300 and stay valid for 12 months. Prices read from solmi.ai/pricing on 2026-11-05, shown as billed monthly.
| Tier | Monthly price | What it includes |
|---|---|---|
| Free | Free | |
| Plus | $35/mo | |
| Creator | $85/mo | |
| Pro | $135/mo |
Key Features
- Song-to-video from an audio file: Audio alone is enough: a video file's soundtrack also works, and the storyboard is written from the beat and lyrics it detects.
- Lip-synced performer from your photo: Add up to three front-facing reference photos and Solmi casts one singer per photo and lip-syncs them to the vocal.
- Shot-level editing in the Studio: Every video stays a project where you can rewrite or regenerate one shot and keep the rest of the edit.
- Lyric detection in 14 languages: Lyrics are detected in 14 languages, including Japanese, Korean and Chinese, and the animated captions can be toggled off.
- Platform-ready aspect ratios: Videos render in 16:9 for YouTube or 9:16 for TikTok, Reels and Shorts, with direct publishing to YouTube and TikTok.
- MCP server and REST API: An MCP server lets Claude, ChatGPT or Cursor create songs and music videos on your account, and a REST API covers scripts.
Pros
- The free tier needs no card and renders a real 20-second preview of your own song before you pay.
- Casting and lip-syncing a singer from up to three photos covers duets, which most song-first rivals do not document.
- Regenerating a single shot costs only that shot, so fixing one bad scene does not burn the whole song's credits.
- Native iOS and Android apps sit alongside the web app, and the vendor says all three have the same features.
Cons
- Output tops out at 1080p, which is thin for a large-screen premiere next to tools that upscale to 4K.
- The terms let Solmi train on your inputs and outputs unless you opt out, a concern for an unreleased master.
- Credits do not map cleanly to minutes: Plus's 2,000 credits are quoted as anywhere from 3 to 7 one-minute videos.
- Free-plan videos are for personal use only, so any monetised release needs a paid plan or a credit pack.
Product Information
- Cloud
- Yes
- Training
- Help Center, Community Discord
Data Handling
- Training-data policy
- Terms section 5.3: inputs and outputs may be used to train Solmi's models unless you opt out in the account data-use setting. Face and voice likeness is used for training only with opt-in.
- Compliance
- Singapore PDPA (privacy policy)
Frequently Asked Questions
What are Solmi's pricing plans in 2026?
Solmi has a free plan and three monthly plans: Plus at $35 for 2,000 credits, Creator at $85 for 5,500 and Pro at $135 for 9,000. Extra credits can also be bought once, without a subscription. Edits and re-renders spend credits, so budget for a few passes per song.
What do you get on Solmi's free tier?
A new web account gets 100 one-time music video credits with no card, enough for one 20-second 480p preview of your song with a watermark. Free song credits refill to 10 a day. Free creations are for personal, non-commercial use only.
What are the best alternatives to Solmi?
Freebeat suits creators who want to generate the song and the video in one place. Neural Frames is the better fit for audio-reactive visuals driven by individual stems. Runway gives shot-by-shot control if you are willing to assemble the edit yourself.
Solmi or Freebeat: which should you pick?
Both are song-first. Solmi's edge is a singer cast from your own photo and lip-synced to the vocal, with output up to 1080p on paid plans. Freebeat lists 720p on its entry plans but bundles music generation models, so pick it if you do not have a finished track yet.
What does it take to start using Solmi?
Sign up on solmi.ai or the iOS or Android app, upload a finished song file and pick a style and aspect ratio. Add a front-facing photo if you want to appear as the singer. Your first preview renders on the free credits in minutes.
Top Alternatives
- Freebeat: Pick Freebeat if you want to make the song and the video under one login.
- Neural Frames: Choose Neural Frames when stem-level, audio-reactive control matters more than a singer on screen.
- Runway: Pick Runway when you want to build each shot yourself from prompts and cut it to the beat.