Guides
Sume guides: the API, Formats, agents, models, media tools and pricing, each answer dated and sourced.
AI generated video disclosure checklist: TikTok, YouTube, Instagram
A pre-publish checklist for labeling AI video on YouTube and TikTok from their own pages, with a note on Instagram, where no page was read for this post.
4K AI image: generate at 4K or upscale a 1K image? The cost
Nano Banana can render at a 4K tier on Sume. Image Upscale 1.0 enlarges an existing image. What each costs per image and what each needs as input.
Why an AI image costs list × 1.25 on Sume: worked examples
Sume's image catalog prices a model at its list price times 1.25, rounded up to a cent. Worked examples for five models and where the number shows.
Am I charged when an AI image generation fails or is canceled?
On Sume's image API a completed image is billed in full and a failed or canceled one is not. How the reserve, capture and refund work, and what a 402 means.
AI image for Instagram 4:5 (1080×1350): what the API returns
Instagram portrait is 4:5, 1080×1350, not 4:3. How to ask Sume's image API for 4:5 and why some models return a native size close to it, not exact pixels.
AI image negative prompt: what to do when the API has no field
Sume's image API has no negative prompt field. Describe what you want instead, edit with a mask, and keep references. The fields that do exist.
AI image generator with readable text: which Sume models to try
Which Sume image models have a vendor claim about text in images, what each vendor says, and the quality setting Sume's docs suggest for dense text.
AI image upscaler 8x: what Sume's upscale API allows
Sume's Image Upscale 1.0 takes an upscale_factor from 1 to 4, so a request for 8 is rejected. What it accepts, costs, and what to do when you need more pixels.
Instrumental AI music by API: how to ask for no vocals
To get instrumental AI music, end the prompt with 'Instrumental, no vocals.' How Lyria 3.5 and Sume's music API handle exclusions.
AI luxury product ad: a cinematic studio commercial by API
A luxury-style product ad can be one API call. Sume's sume-cinematic-studio-commercial Format makes a studio commercial from a packshot; the call and limits.
AI music generator API in Python: prompt to MP3 file
A short Python script that sends a prompt to Sume's Music Router, polls the job and saves the audio. Runs with httpx and asyncio.
AI music BPM control: set the tempo in the prompt
There is no BPM parameter in Lyria 3.5 or Sume's music API. Write the tempo as a number in the prompt, then check the result. How, and what to expect.
AI music negative prompt: why the API returns 400
Sume's music API rejects a non-empty negative_prompt with HTTP 400. What the error looks like and how to write exclusions in the prompt instead.
AI music stems: does the API return separate tracks?
Sume's music API returns one audio file, not stems. What Lyria 3.5's documentation covers, and how to get parts you can mix separately.
Does Lyria 3.5 music carry a SynthID watermark?
Google says all Lyria 3.5 audio includes a SynthID watermark. What that means for tracks made through an API, and what Sume's docs do and do not say.
AI music reference artist song blocked: write the brief without one
ElevenLabs blocks downloads of tracks that reference other artists' songs. Describe the sound by emotion, genre, tempo and instruments, as Sume's brief does.
AI music API webhook: get a callback when the track is ready
Submit a music job with mode webhook and a public HTTPS URL, verify the signed callback, and keep polling as a backup. Headers, events and retries.
AI product photos: cutout, new scene, upscale, and the cost per SKU
A three-call product photo pipeline on Sume: remove the background, generate a scene from the cutout, upscale the winner. What each call costs per SKU.
AI product photo: change the background to a lifestyle scene
Send the white-background product shot as a reference with an edit prompt and aspect_ratio auto, and Sume returns a new scene. A mask keeps the product fixed.
AI profile picture generator from a photo
Turn one photo into a square profile picture with Sume's image API: send the photo as a reference, ask for a 1:1 result, then optionally cut out the background.
AI SDK 7 createMCPClient: connect a remote HTTP MCP server
In AI SDK 7, createMCPClient takes an http transport with a url and headers. Point it at https://mcp.sume.com/mcp, load client.tools(), and close in finally.
AI SDK 7 isStepCount stopWhen: cap a paid tool loop
stopWhen: isStepCount(5) limits AI SDK steps, not spend. For paid Sume MCP tools, pair it with max_spend_usd, dry_run and an idempotency_key.
AI SDK MCP client authorization bearer header: what to send Sume
Set headers.Authorization to Bearer plus your Sume API key on the AI SDK http transport, and send only one credential header. An OAuth token is not an API key.
AI song structure tags: [Verse], [Chorus] and timestamps
How to structure an AI song prompt for Lyria 3.5 with section tags and timestamps, and how the same prompt goes through Sume's Music Router.
Sound effects for video AI: write the sounds into the clip prompt
With no sound-effects endpoint, video effects come from models that generate audio with the clip. Check the generate_audio flag, then describe each sound.
AI sound effects for video: what Sume lists and what it does not
Sume does not list a dedicated sound-effects model. Sound for video comes from video models with generate_audio, the music route, or your files on the Timeline.
AI UGC video generator API: one call to a close-camera Format
A UGC-style video generator can be one API call. Sume's sume-close-camera-ugc Format makes a handheld selfie-style video; see the request and the receipt.
24fps vs 30fps AI video: which output.fps to set on Sume
Should an AI video be 24 or 30 fps? On Sume's Timeline, leave output.fps unset to follow your sources; set 24, 25, 30 or 60 only when a delivery spec needs it.
AI video API cost calculator: price a clip from pricing_skus
Build a video cost calculator from Sume's pricing_skus: per-second, per-resolution and per-token rates, a short Python function, and worked examples.
AI video multiple faces consistency: Luma's 8-face limit, Sume's caps
Luma Ray3.2 lists facial performance tracking for up to 8 faces. Sume has no such setting; the limit is how many reference images each video model accepts.
Why AI video models do not lip sync to a voiceover, and what to use
Sume docs: video models do not lip-sync to generated speech or a voice-over. For a talking face use Avatar video or the lip-sync endpoint (still plus audio).
Reframe an AI video to 9:16: regenerate, crop or fit?
Luma lists enhanced reframe for Ray3.2. On Sume, a 9:16 frame comes from asking for it, cropping with video filter, or fitting in Timeline.
AI voiceover for short videos: 9:16 with voice and music
Make an AI voiceover for a Reel, Short or TikTok: script to speech, clips on the beat of the voice, optional music bed, one vertical MP4.
Amazon Fire tablet video ad specs: 15 s max, 16:9 or 16:10, audio
Amazon's templated Fire tablet video ads allow 15 seconds, 1280x720 or larger, 16:9 or 16:10, .mp4 under 500 MB with AAC audio at 48 kHz. Sume fit.
Amazon Fire TV feature rotator video specs: 1920x1080, -24 LKFS
Amazon lists the Fire TV feature rotator video at 5–21 seconds, 16:9, 1920x1080, no borders, and loudness of -24 LKFS. What Sume can and cannot set.
Amazon Fire TV inline video ad specs: 10 to 30 s, 16:9, 960x540 min
Amazon Ads lists Fire TV inline video at 10–30 seconds, 16:9, 960x540 minimum, 500 MB, 2,000 kbps and audio at 128 kbps or higher. Sume can make the picture.
API audit log: what you can trace on Sume's media API
Sume's docs describe no audit-log endpoint. Job records, the usage ledger and request ids give a trail, but not which API key made a call.
API changelog: where Sume lists changes, and how to read it
Sume keeps a public changelog, newest first, with a version, a date and short notes. It does not mark breaking changes, so also watch the retiring list.
API Gateway 29 second timeout: calling a video API from Lambda
API Gateway REST APIs cut integrations at 29 seconds by default. Submit the video job, return 202, and take the result on a callback.
API spend limit: can you set a monthly budget on Sume?
Sume's docs describe no monthly spend limit. The controls they do describe are the wallet balance, a per-run spend cap, and plan concurrency.
App Store screenshot size (1290×2796) and AI images
Apple lists 1290×2796 for 6.9-inch iPhone screenshots. Sume's ChatGPT Image 2.5 cannot output it exactly, so generate the same shape larger and resize.
Article URL to video AI: turn a blog post link into a video
An article URL to video tool makes a short video from a post. On Sume send the article text in input and its images as attachments; the link alone is not read.
Refresh old YouTube thumbnails with AI: a batch method
YouTube says Ask Studio will suggest refreshed thumbnails. To refresh a back catalog yourself, send each old thumbnail to an image API as a reference.
Ask YouTube shopping: get my product recommended, what to make
YouTube is adding a product recommendation guide to Ask YouTube. It has published no ranking rules, so prepare product videos that answer buyer questions.
Extract 16 kHz mono audio from a video for speech-to-text
Set sample_rate 16000 and channels mono on Sume's audio detach to get the speech-to-text shape from a video. Options, the 900 second cap and the price.
detach_source_has_no_audio: check for an audio track first
Audio detach fails with detach_source_has_no_audio on a silent clip. Probe has_audio with a video inspect that skips frames, then detach only clips with sound.
AI avatar video aspect ratio: 9:16, 1:1, 3:4, 4:3, 16:9
Sume's avatar video accepts five aspect ratios and defaults to 9:16 at 720p. Which ones are portrait, square or landscape, where to set it, and when it locks.
AI video captions failed: is the video still usable?
On a Sume avatar video, a failed caption stage does not fail the job: the clean video_url still arrives with captions.status set to failed. What to do next.