Blog
Sume blog: agent-native video generation, Avatar 1.0 orchestration, and production APIs for builders and coding agents.
Does Sume have a real-time avatar API? No, here is what it has instead
Sume has no live avatar session. It has async avatar jobs: create an avatar, render a talking video, or lip-sync a still to audio. Routes, limits and prices.
How do I make AI announcer intros for conference speakers?
Ten 25-second speaker intros with one shared sting cost about 43 cents on Sume: 2 cents of TTS and a 1-cent join each, plus a $0.125 track made once.
How do I make a safety briefing audio in three languages with TTS?
One 1,800-character briefing in English, Spanish and Polish is three TTS jobs: 9 cents each, 27 cents in all on Sume. Set language on every job and review it.
Fade in and out on a Timeline render: output fade seconds 0 to 5
Set output.fade_in_seconds and fade_out_seconds (0 to 5 s, sum within the render length). The music bed has its own fade_out_seconds, up to 10.
Fast-cut Shorts in Timeline: eight chained fades, then a hard cut
Timeline refuses more than 8 adjacent fades with too_many_chained_transitions. Transitions must be 1 s or less and half the shorter neighbour. How to plan cuts.
Fix extra fingers in an AI image: a masked GPT Image 2.5 edit
Fix a bad hand in an AI image with a mask_url edit on openai/gpt-image-2.5: mask only the hand, keep the rest, and budget about $0.0835 per try at high.
Game NPC dialogue with AI avatar clips: pre-render lines, not live
For a game or interactive story, 120 pre-rendered 5-second NPC lines cost $147.00 on Sume's plus tier. When clips are enough and when you need a live character.
Get transcript text from a captioned video: caption jobs return none
A Sume caption job returns the burned video, not the transcript. For text and word times run STT or video inspect with transcribe. Prices and a recipe.
Why a vertical-only YouTube ad may not serve, and the 16:9 fix
Google says campaigns with only vertical videos may not serve in YouTube in-feed. Render a 16:9 twin of your 9:16 ad from one audio spine with Sume Timeline.
Can a 4:5 or 2:3 video run in Google Ads? What the specs allow
Google Ads lists 16:9, 9:16 and 1:1 as primary and says ratios in between are allowed; a spec page also names 4:3, 2:3 and 4:5. Exact sizes for Sume Timeline.
OpenAI names Flare or Sunburst to replace gpt-image-1: which to pin?
OpenAI's deprecations page lists gpt-image-2.5-sunburst or -flare for gpt-image-1. Sume serves both at one price, so the choice is quality. A script to A/B.
GPT Image 2.5 at 4:5 lists 12-14% below 1:1 at the same quality
On Sume, GPT Image 2.5 at 4:5 (1024x1280) lists 12 to 14 percent below a 1024x1024 square at every quality tier. The five-tier table, with billed prices.
Headshot background to neutral grey: cutout plus Pillow or an AI edit
Swap a headshot background for grey: Sume RMBG at $0.0225 plus a Pillow composite keeps the face pixels exact; an Ideogram 4.5 edit costs $0.075 or more.
Hook line in instruction or input? Sume Format run for ad variants
Put the hook text in the instruction field, and put product data in input. Rules, limits and errors for varying an ad hook across Sume Format runs.
A 3-minute Timeline render: poll job status, don't sleep a fixed time
Timeline renders are async by default. Submit with an Idempotency-Key, poll /v1/jobs/:id/status until terminal, then read /result. Cost: 3 minutes is $0.30.
How many 6-second clips fill a 3-minute Short, Reel or TikTok ad?
A 3-minute Short needs 30 six-second clips, a 20-minute Reel 200, a 10-minute TikTok ad 100. Slot limits, chunked renders and Timeline cost for each.
Image edit price sheet on Sume: Ideogram 4.5, GPT 2.5, cutout, upscale
What image editing costs on Sume, October 2026: Ideogram 4.5 $0.0375 to $0.275, GPT Image 2.5 by quality, RMBG $0.0225, Upscale $0.20, and chained pipelines.
Instagram Reels templates need music and 3 clips: prepare them
Instagram says templates need a reel with music and at least 3 clips. Render one master with Sume Timeline, then cut three clips with video trim at $0.02 each.
Instagram says upload the highest resolution possible: what's the max?
Instagram's creator page says upload the highest resolution possible but gives no pixel number. Sume Timeline's ceiling is 2160 per edge, so 1214x2160 vertical.
Is there a Runway Ads API? What Runway's own pages list
Runway's home page lists Creative, Dev and Robotics; its dev docs name ad recipes. What we could confirm, and the Sume route for ad variants.
How do I keep a product label exact in an AI product video?
Animate the product photo, then overlay your real label still with Timeline compose: $0.63 for a 5-second clip plus $0.02 for the overlay on Sume.
How do I add a listen-to-this-page audio version with TTS?
Turn each article into an audio file with one async TTS job per page: a 9,000-character article costs 43 cents on Sume. What it does not replace.
Live stream avatar: queue pre-rendered reaction clips instead
For a stream overlay, render ten short Sume avatar clips ahead of time and trigger them from chat events. Ten 6-second clips cost $14.70 on plus.
MAI-Transcribe-2-Streaming intro rate runs to year end: budget plan
MAI-Transcribe-2-Streaming is $0.54 an hour through year end. How to budget transcription hours against Sume STT at $0.60 an hour (one cent a minute).
Meta Feed video ad spec checklist before uploading an AI clip
Meta's Facebook Feed video spec: 4:5, 1440 x 1800, up to 4 GB, MP4 or MOV, H.264 with AAC audio. A checklist, and the Sume Timeline output size that meets it.
Model an AI generation job as a state machine in your database
A schema and update rule for tracking Sume jobs: five statuses, sticky terminal states, a separate webhook delivery column, and the idempotency key on the row.
Nano Banana Pro or GPT Image 2.5 xhigh and max at 1K: price on Sume
Nano Banana Pro at 1K bills $0.1875 on Sume; ChatGPT Image 2.5 xhigh bills $0.1171 and max $0.2635 at 1024x1024. Ratios, references and the order of the three.
What does one hour of AI narration cost on Sume, MAI and ElevenLabs?
One hour of finished narration is about 45,000 characters: $2.16 on Sume in three jobs, against $0.99 on MAI-Voice-2.1. Rates read 2026-10-07.
Smallest vertical video size that passes Google Ads and TikTok
Google lists 720x1280 as the vertical minimum, TikTok in-feed 540x960 and its app bundle 720x1280. One Timeline output size clears them all, with the table.
One voiceover, two ratios: render 9:16 and 1:1 from one audio file
Reuse one voiceover for a 1080x1920 and a 1080x1080 cut in Sume Timeline. Two renders at $0.10 per output minute each; only output width and height change.
How to make a pep talk audio track with music: voice, bed and render
A 90-second pep talk with a music bed is one TTS job, one Music Router job and one render: about $0.39 on Sume. Suno Speech beta does it in one pass.
How do I turn one podcast episode into five quote clips for social?
Transcribe the episode in 10-minute chunks, split five quotes out, put the cover still under each and burn captions: $1.80 for a 28-minute episode on Sume.
Poll hundreds of AI jobs without a thundering herd: jitter and budgets
Poll many Sume jobs without synchronized bursts: jitter, next_poll_after_seconds, per-plan read budgets, and the math on how much polling a plan can absorb.
Portuguese speech to text API: Sume STT language_code pt or pt-BR
Transcribe Portuguese audio with Sume STT using language_code pt or pt-BR, then check the reported language and word times. $0.01 per audio minute.
Post-call recap video with an AI avatar for prospects: script and cost
After a sales call, send a 30-second recap clip from a Sume avatar. Script structure, cost at standard, plus and max, and how to keep it honest and reviewed.
Probe a finished video before upload: duration, size and aspect
Run video inspect with frames false to read a render's duration, size and frame rate before posting. Check it against the 3-minute Shorts limit.
Product photo to a 6-second vertical ad: a still is a static hold
A still in a Timeline video slot is held, not animated: motion is accepted but ignored with a motion_ignored warning. Default 1080x1920, $0.10 for 6 seconds.
Real-time or async avatar: six questions to ask before you choose
Need an avatar that talks live, or a finished clip? Six yes/no questions decide it, with what Sume's async avatar jobs cover and what they do not.
Reconcile Sume jobs after a deploy or outage: poll what is open
After downtime, read status for every job your own table still shows as open, honor terminal and result_ready, and never resubmit. Python with sqlite.
Redact faces and license plates: Pillow first, AI edit only to replace
For redaction use Pillow boxes you control; use an AI mask edit on openai/gpt-image-2.5 only to replace a plate or face, from $0.0094 per image on Sume.
Remove a date stamp from a scanned photo: crop first, then a mask edit
Remove an orange date stamp from a scanned family photo: crop it for free in Pillow, or mask it and edit with openai/gpt-image-2.5 from $0.0094 an image.
Remove an object from a photo without a mask: Ideogram 4.5 prompt edit
Remove a bin, cable or stray person from a photo with no mask: send it to ideogram/ideogram-v4.5 and name the object. $0.0375 to $0.275 per image on Sume.
Replace a dull sky in a property photo: mask only the sky
Swap a grey sky for blue in a property photo: build a sky mask in Pillow and send a mask_url edit to gpt-image-2.5. $0.021 at medium, $2.10 for 100 photos.
How much does a holiday catering promo reel with six dishes cost?
Six dish photos animated 5 seconds each, a voice, music and price captions: $4.235 for a 30-second catering reel on Sume at 720p, or $1.14 for the drafts.
How do I make a Thanksgiving takeout menu video for my restaurant?
Animate two of your own food photos with Gemini Omni Flash 1.1, add a 380-character voiceover and burn the order-by date in: $1.945 on Sume, in 12 seconds.
Move burned-in captions up or down: restyle with source_caption_id
Change caption placement with design.placement.anchor_ratio and restyle with source_caption_id: the same word timings, no second speech-to-text run.
Runway Dev usage exception request vs Sume 429 and 402
Runway Dev says higher usage needs an exception request. On Sume a burst hits 429 or 402. What each means for a batch of ad variants and how to retry.
How do I make a self-guided walking tour audio with TTS?
A six-stop walking tour at 700 characters a stop is six TTS jobs of 4 cents each, 24 cents on Sume, about 0.9 MB per stop as default mp3.