Solmi is an AI music video director and generator that turns a finished song into a storyboarded, lip-synced music video. It listens to your track — lyrics, sections, beat — writes a creative brief, drafts a shot-by-shot storyboard with recurring characters, and only renders after you've approved the plan. Lyric detection and lip-sync work in 14 languages, every new account gets a free 20-second preview, and full-length renders start at $35/month. Give notes in plain language, regenerate any single shot, and export the finished video for YouTube, TikTok, or Reels.
What Solmi's AI music video director accepts, produces, and costs, in one place.
| Spec | Solmi |
|---|---|
| Input formats | MP3 or WAV upload; JPG, PNG, or WebP photos for performer casting |
| Song sources | Your own upload, including a song downloaded from Suno |
| Lyric detection & lip-sync languages | 14: English, Spanish, French, German, Italian, Portuguese, Dutch, Polish, Turkish, Japanese, Korean, Chinese, Indonesian, Vietnamese |
| Storyboard editing | Plain-language rewrite per shot — reorder, rewrite, or regenerate any single shot before rendering |
| Performer casting | Cast from your own photos; each character gets an approved reference image kept consistent across scenes |
| Aspect ratios | 16:9 (YouTube) or 9:16 (TikTok, Reels, Shorts), picked when you start the project — MP4 |
| Max resolution | 1080p on paid plans (720p also available); 480p free preview |
| Free tier | One-time 100 music-video credits = a single 20-second 480p preview with watermark; song credits refill to 10/day; no credit card |
| Plans | Plus $35/month for 2,000 credits (~3–7 one-minute music videos); Creator $85/month for 5,500; Pro $135/month for 9,000 — all 720p & 1080p, no watermark, commercial use |
| Platforms | Web, iOS (App Store), Android (Google Play) — one account, one credit balance |
| Typical render time | Brief and storyboard in about a minute; finished music video in 3 to 10 minutes |
Solmi is credit-based. Every new account starts with a one-time 100 music-video credits, enough for a single 20-second 480p preview of your song, plus song credits that refill to 10 per day — no credit card required. In the Director, each step (song analysis, brief, storyboard frames, and the final renders) shows its credit price before you run it, and regenerating one shot only costs that shot. Full-length renders at 720p or 1080p with no watermark come with the paid plans: Plus at $35/month for 2,000 credits (roughly 3 to 7 one-minute music videos), Creator at $85/month for 5,500, and Pro at $135/month for 9,000. Paid credits reset each billing cycle and do not roll over.
Solmi is built as an AI music video director: it analyzes your song's lyrics, structure, and mood, writes a creative brief, drafts a shot-by-shot storyboard with recurring characters, and then renders and lip-syncs the finished video — all in one workflow. Tools like Neural Frames offer deep frame-level prompting but you direct each scene manually, and most one-click generators (freebeat, Plazmapunk) skip the storyboard stage entirely.
Solmi generates an editable storyboard automatically before any video is rendered: every shot has a frame preview, a scene description, and assigned characters, and you can reorder, rewrite, or regenerate individual shots. Most AI music video generators go straight from song to final video with no reviewable shot-planning step in between.
In Solmi's Director, each character and location gets a reference image you approve before rendering. If a storyboard frame drifts from the approved reference, you can regenerate just that shot — you're always shown the credit cost before anything renders, so iterating on single shots stays cheap compared to re-rendering the whole video.
Solmi drafts a complete brief and storyboard for a full-length song in about a minute, and you can start reviewing shots immediately while the rest generate. Rendering the final video takes minutes, not hours, because shots render in parallel and you only re-render the ones you change.
Yes. Solmi's Director is chat-based: tell it things like "make the chorus neon-lit" or "put the singer on a rooftop at night" and it updates the brief and the affected storyboard shots. You approve changes before anything is re-rendered.
Yes. New accounts receive 100 starter music-video credits to try the Director workflow. Song analysis, creative generation, storyboard images, and finished video renders use credits as you move through the project.