A Musician Made a Music Video With Suno, ChatGPT and Kling for 2,000 Credits and a Week of ChatGPT Plus Limit
Musician kay-n made the video "Platye" without a film crew. Suno generated the song, ChatGPT drew the characters and the storyboard, and Kling 4.0 produced the video with lip-sync. The spend was a week of the ChatGPT Plus subscription limit and about 2,000 credits, which are Kling's internal currency. The footage was generated at 720p and then upscaled in Topaz Video to keep costs down. For an indie artist, this is a ready-made budget: storyboard in the cheaper ChatGPT step, video last.
AI-processed from TG: @cgevent (Метаверсище и ИИще); edited by Hamidun News
Music videos used to require a film crew, locations and editing. Now an independent artist can put one together from a few subscriptions, and one such case was shown in the channel "Metaverseshche i IIshche". The musician under the nickname kay-n made the video "Platye" for his project "Aramboll": Suno (a service that generates music from a description) wrote the song, ChatGPT created the characters and the storyboard, and Kling 4.
0 (a video generation service) produced the footage and the lip-sync with the vocals. The cost was roughly a week of the ChatGPT Plus subscription limit and about 2,000 credits. Credits are Kling's internal currency: each generation deducts part of the balance.
The author disclosed the spend openly, and figures like these are rarely published.
The project started two years ago. The author wanted to understand what neural networks can do and learn to work with them. Over this time, two full albums and six music videos came out, with "Platye" the latest of them. This is not a one-off experiment: the author already has a lot of trial and error behind him.
How the video was assembled
Each tool in the chain has its own role.
Suno made the song. It comes first, because it sets the length of the video, the mood and the rhythm of the visuals.
ChatGPT drew the characters and put together the storyboard, that is, the sequence of shots for the future video. This took roughly a week of the Plus subscription limit. Mistakes at this stage are cheap: an image can be regenerated without spending video credits.
Kling 4.0 produced the video itself and the lip-sync. This consumed about 2,000 credits, the most expensive part of the work.
Topaz Video increased the resolution of the finished clips. This program was used to save money, more on that below.
How the author saved money
The author generated the video at 720p rather than at maximum quality. The higher the resolution, the more credits each clip costs. So he paid for a cheap draft and added sharpness afterward in Topaz Video. The post does not say how much this reduced the spend. The approach itself is clear: inexpensive generation first, then a separate enhancement step.
The order of work also protects the budget. First the song, then the characters and the storyboard in ChatGPT, where a mistake costs little. The video model comes in last, once it is clear what to shoot. If you run video generation blindly from the start, the credits go on random attempts.
What this means in practice
For an independent musician or a small studio, the post gives a budget benchmark. A video like this costs a monthly ChatGPT subscription, a credit pack for a video service and one upscaler. That is not zero, but it is also not a shooting day with equipment. For a product manager or a startup founder, it is an example of building creative work from a combination of narrow services rather than looking for one universal tool.
There are limitations too. The post describes a single case. It does not show how many attempts each shot required, how long the assembly took, or how much manual editing happened off-camera. The figures of 2,000 credits and a week of the Plus limit apply to the plans the author used. Other users may see different costs, and service limits and credit prices change, so the budget is best calculated against current terms. Each viewer will judge the quality of the result for themselves: the video can be watched in the channel.
Three rules can be taken from this case. First, build the plot and characters in cheap text and image tools. Then generate video at a moderate resolution. Finally, enhance the result with a separate tool. While the models write the song, draw the characters and move the lips, the author's role stays the same: he sets the concept and selects the successful results.
- Nov 21, 2026Apple Is Building an AI Agent and Making It Harder for Outsiders to Get Access: Permission Now Takes 14 Steps, Not 4
- Nov 21, 2026Developer Skips Pinecone: Searching 72,450 Code Chunks in 50 ms
- Nov 21, 2026Gemini 4 Neon shrank the TS Rust binary from 47.44 to 7.97 MB in under 10 hours
- Nov 21, 2026FDE job postings grew 4.6x in a year: what a forward deployed engineer actually does at the client's site
Key Questions Answered
How the video was assembled?
Each tool in the chain has its own role. Suno made the song. It comes first, because it sets the length of the video, the mood and the rhythm of the visuals. ChatGPT drew the characters and put together the storyboard, that is, the sequence of shots for the future video. This took roughly a week of the Plus subscription limit. Mistakes at this stage are cheap: an image can be regenerated without spending video credits. Kling 4.0 produced the video itself and the lip-sync.
How the author saved money?
The author generated the video at 720p rather than at maximum quality. The higher the resolution, the more credits each clip costs. So he paid for a cheap draft and added sharpness afterward in Topaz Video. The post does not say how much this reduced the spend. The approach itself is clear: inexpensive generation first, then a separate enhancement step. The order of work also protects the budget. First the song, then the characters and the storyboard in ChatGPT, where a mistake costs little.
What this means in practice?
For an independent musician or a small studio, the post gives a budget benchmark. A video like this costs a monthly ChatGPT subscription, a credit pack for a video service and one upscaler. That is not zero, but it is also not a shooting day with equipment. For a product manager or a startup founder, it is an example of building creative work from a combination of narrow services rather than looking for one universal tool. There are limitations too. The post describes a single case.
Ship AI Faster with Founders, Engineers & Weekly Architecture Reviews
Inside Vibe Club (hamidun.com/club) — weekly live masterminds with Zhemal Khamidun, Tech Radar, full LMS access, and 370+ production Claude Code skills.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.