Product Hunt launch video: the YouTube link and gallery images
Product Hunt takes video only as a YouTube link and wants 2+ gallery images near 1270x760. Make the video with Sume Timeline and images at a size GPT accepts.

For a Product Hunt launch you need three assets: a video that lives on YouTube, at least two gallery images, and a square thumbnail. Sume can make the video file and the images. It cannot post to YouTube or Product Hunt for you, so the video is a file you upload to YouTube yourself and then paste the link. Gallery images should be requested at 1280x768, because Product Hunt's recommended 1270x760 is not a size the GPT image models accept.
What does Product Hunt accept?
Everything here is from Product Hunt's how to post a product page.
| Asset | Rule on the page |
|---|---|
| Gallery | 2 or more images before it is viewable; 1270x760 recommended; reorder by drag and drop |
| Video | Only YouTube links; full URLs, not shortened; not private before launch |
| Thumbnail | Best square, 240x240 recommended; animated GIFs allowed under 3 MB; strobing and quick cuts discouraged |
| Description | Within 260 characters |
How do I make the video?
Build the demo as a Timeline render: one narration file as the audio spine, screen recordings or stills as ordered slots. A Timeline result is a media.sume.com MP4 you download from the job result and upload to YouTube. Timeline output is 1080x1920 by default, so set output.width 1920 and output.height 1080 for a landscape launch video. Do not leave it private on YouTube, since Product Hunt says the video must not be set to private before launch. Timeline bills $0.10 per output minute.
curl -X POST https://api.sume.com/v1/timeline-1.0/render \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: ph-launch-001" \
-d '{
"audio": { "url": "https://media.sume.com/artifacts/artf_vo/voice.wav", "duration_seconds": 45 },
"output": { "width": 1920, "height": 1080 },
"video": [
{ "source_url": "https://media.sume.com/artifacts/artf_rec/problem.mp4", "start": 0, "duration": 15 },
{ "source_url": "https://media.sume.com/artifacts/artf_rec/product.mp4", "start": 15, "duration": 30,
"transition": { "type": "fade", "duration": 0.25 } }
]
}'Why 1280x768 and not 1270x760?
Per the image docs, ChatGPT Image 2.5 takes custom image_size values only when both edges are multiples of 16, the maximum edge is 3840, the aspect ratio is at most 3:1, and the pixel count is between 655,360 and 8,294,400. 1270 and 760 are not multiples of 16, so a 1270x760 request is outside the rule. 1280x768 is (80 x 16 by 48 x 16), at 983,040 pixels, and its 5:3 ratio is within a hair of Product Hunt's. Check the model's list in GET /v1/images/models before you send, as the catalog is the source of truth.
curl -X POST https://api.sume.com/v1/images \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: ph-gallery-001" \
-d '{
"model": "openai/gpt-image-2.5",
"prompt": "Clean product screenshot mockup, dashboard on a laptop, soft daylight",
"image_size": "1280x768"
}'What about the 240x240 thumbnail?
240x240 is 57,600 pixels, far under the 655,360-pixel floor, so you cannot request it directly. Generate a square at 1024x1024, then scale it down in any image tool. Product Hunt says to avoid strobing and quick cuts in animated thumbnails, so if you use a motion thumbnail, keep it a slow loop under 3 MB.
What should I not claim?
Gallery images are product shots, so show the real product screen. Use generated images for context, not for features you do not have. Product Hunt's description limit is 260 characters, and none of these tools write it for you; draft it in your own words.
Sources
Related posts
More in Use cases
- Product launch week content budget: hero video, 5 Reels, 3 carousels
A launch week with research, one 30 s hero video, five 15 s Reels and three carousels costs about $32 on Sume. Itemised, with the levers that move it.
- Product review Shorts per SKU: vary the focus, not just the name
YouTube allows a review channel when each video has a distinct focus and rejects minimal variation. How to plan a batch of per-product Shorts for that test.
- QA every episode's cliffhanger line with a video inspect transcript
Before you post episode 5, transcribe the render with POST /v1/video-inspect and check the last sentence. $0.01 per audio minute, plus a short Python gate.
- Qwen Image Max backdrop plates for product composites
Generate empty lifestyle backdrops with Qwen Image Max, then place the real product photo: vendor facts, price, and Sume's text-only catalog entry.
Written by Sume