Blog
Sume blog: agent-native video generation, Avatar 1.0 orchestration, and production APIs for builders and coding agents.
HeyGen's X-RateLimit-Scope header vs Sume's error.details.scope
HeyGen's October 2026 429s add a scope header. Sume names the bucket in error.details.scope and sends ratelimit-* headers; one parser can read both.
Sume hosted MCP: API key or OAuth for an overnight agent job?
OAuth fits a person at a keyboard; an API key fits an unattended agent. How Sume hosted MCP treats each, plus the guardrails to set before you leave it running.
HVAC seasonal promo: 12 neighborhood versions in one Format bulk run
A heating and cooling company can queue a furnace-tune-up promo for 12 neighborhoods in one bulk run: concurrency, per-item input, and a spend cap per run.
Image burst after a model launch: 429 queue_full vs 503 retry plan
Launch week means batches. On Sume, 429 rate_limited, 429 queue_full and 503 provider_capacity_exceeded each need a different retry, plus idempotency keys.
image_size, aspect_ratio or size: which field wins on Sume
Sume's image API has three size fields. image_size beats aspect_ratio, size takes only a tier, and 4:5 is 1080x1350 portrait. Examples for GPT and Nano Banana.
Jewelry ring close-up video at 1080p: Omni on Sume vs Veo Lite
A sparkle close-up for a ring at 1080p. Omni on Sume lets you pick 3 to 10 seconds; Google says Veo 3.1 Lite 1080p is 8 seconds only. Prompt, prices and checks.
Kling 3.0 Element Binding: a face close-up, and one still on Sume
Kling says 3.0 Motion Control wants a clear facial close-up for Element Binding. Sume's route takes one still or a ready avatar. What that means for faces.
Kling motion control: 3 s minimum on Kling, 1 s field on Sume
Kling says action videos run 3 to 30 seconds. Sume's duration_seconds accepts 1 to 30. Why to keep the reference at 3 seconds or more, and what it costs.
Korean BBQ restaurant reel: korean-ad captions on spoken Hangul
Caption a restaurant reel with Korean speech using Sume's korean-ad style, language ko, and why slam or punch fails with Hangul. Request body and font choices.
Lip-sync and motion control cost per clip: H3 Max, Fabric, Kling
Sume prices H3 Max lip-sync at $0.0625 to $0.20 per audio second, Fabric at $0.1875 and Kling 3.0 Motion Control at $0.1575. A 10-second clip is $0.63 to $2.00.
List Sume image models that take references or masks with jq
New image models land weekly. One curl and one jq filter on GET /v1/images/models show which ids accept input_references, mask_url or background, and how many.
Live avatar mistakes vs rendered avatar retakes: where review fits
A live avatar can say the wrong thing to a customer in real time; a rendered Sume avatar clip can be reviewed and retaken first. Where each puts the checkpoint.
Lofi rain-on-window loop with sound: Gemini Omni 10 seconds
Make a 10-second lofi rainy window clip with Gemini Omni on Sume. Ambient audio wording, one image as first and last frame for a loop, and the cost to draft.
Mortgage broker weekly rate update video: avatar, captions, review
A weekly 30-second rate update clip for a mortgage broker: avatar script, inline captions, scheduled Action, and a human compliance check before posting.
Music 1.0 is retiring: keep old calls, move new code to the router
Sume Music 1.0 routes still work and keep job.model sume/music-1.0, but every request now resolves through the Music Router. What to change and when.
AI music cost per track: Sume Music vs Lyria 3.5 and ACE Step
Sume Music is a flat $0.125 per generation ($12.50 per 100). Google lists Lyria 3.5 at $0.08 per song ($8.00); fal lists ACE Step at $0.0002 per second.
Nano Banana 2.1 prices: Google's 1K and 4K list vs Sume's cost
Google launched Nano Banana 2.1 on Oct 6 at $0.0336 per 1K image and $0.0756 at 4K. Here is how to read what Sume bills for google/nano-banana-2.1.
Nano Banana 2.1 mask editing: Sume's mask_url is GPT-only
Google's Nano Banana 2.1 adds mask-based editing. On Sume, the mask_url field is accepted only by ChatGPT Image 2.5, so pick the model by the edit you need.
New business logo sting for reels: sume-logo-motion-design end card
Animate a supplied logo into a short ident for reels and ads with Sume's logo-motion Format, then attach it to every clip as an end card. Inputs and checks.
Next.js route handler: submit a Seedance 2.5 job, keep the key private
Keep your Sume API key on the server: a Next.js route handler that submits a seedance-2.5 job and returns the job id, plus a second route that polls it.
No-code HTTP step: submit a Wan 3.0 job and get a callback_url
Any automation tool with an HTTP request step and a webhook trigger can run wan-3.0 on Sume. The request body, the HTTPS callback_url, and what to store.
Bumper and non-skippable ads bill per impression: cut 6, 15, 30 s
Google bills bumper and non-skippable in-stream per impression (Target CPM). Length is a creative call: cut 6, 15, and 30 second versions with video-trim.
One 58-second vertical master that fits Shorts, Reels, and TikTok ads
A 58-second 1080x1920 master fits YouTube Shorts, the Reels 3-minute line, TikTok reservation and TopView, and Google's Shorts-ad guide. Render it and verify.
Online course trailer: avatar intro plus b-roll in one timeline render
Assemble a 30-second course trailer: an avatar intro, generated b-roll clips, and one audio spine rendered into a single MP4 with Sume's Timeline 1.0.
Personal trainer class promo: avatar hook, silent demo beat, CTA
Build an 11-second class promo for a fitness studio from three ordered scenes on one avatar, with a silence beat for the demo. Limits and request body included.
Photographer highlight reel: trim three clips, render one timeline
A freelance photographer or videographer can cut three client clips with video-trim and join them with a music bed in Timeline 1.0. Rates, limits, and rights.
Plan in Claude Code over MCP, then hand the batch to bulk runs
Use Claude Code with Sume's hosted MCP to prototype one video and read costs, then run the real batch from your backend with Format bulk runs and a spend cap.
Port a Kling 3 request to Sume motion control: three fields to remove
Sume Kling 3.0 Motion Control has a strict body. Moving from a Video Router kling-3 request means dropping model and reference urls and adding two fields.
Post-production cost for 100 AI clips: trim, caption, upscale
Trim is $0.02, captions $0.20, upscale $0.009 per second. Trim, caption and upscale 100 fifteen-second clips for $35.50 on Sume, then assemble them for $2.50.
POV hiking video prompt for Gemini Omni: handheld, breath, steps
Write a first-person POV trail clip for Gemini Omni: camera wording, body sounds, a 10-second 9:16 Sume request and its cost at 720p and 1080p.
Sume Pro, Startup or Scale: real cost per 10-second clip
A 10-second 768p clip is $1.00 of usage. Plan fees turn that into $1.00, $0.86 or $0.50 per clip if you use the full included amount. Break-evens inside.
Product turntable video from one photo with Gemini Omni
Ask Gemini Omni for a slow orbit around a product from a single photo, then check the back of the object. Sume request, a fallback, and 8-second prices.
Can I submit 100 AI video jobs at once? Queue limits by plan
Accepted capacity is slots plus queue: 6 on Free, 24 on Pro, 48 on Startup, 120 on Scale. Submit 100 at once and 94, 76, 52 or 0 get 429 queue_full.
Qwen-Image 2.1 subject extraction vs Sume's RMBG 1.0 cutout
Qwen-Image 2.1 lists subject extraction and native RGBA. For a product cutout on Sume, use the RMBG 1.0 background-removal route, which returns PNG with alpha.
Ruby Net::HTTP: submit a Wan 3.0 video job and poll until it finishes
A standard-library Ruby script that posts to /v1/videos with wan-3.0, polls the polling_url every 30 seconds and stops on completed, failed or cancelled.
Runway Act-Two at 5 credits a second vs Kling motion control on Sume
Runway's developer page lists Act-Two at 5 credits a second, or $0.05. Sume's Kling 3.0 motion control is $0.1575 a second. What each price buys, in dollars.
Runway API credits to dollars: cost per second and per 100 clips
Runway's API credit is $0.01, so WAN3 is $0.05 to $0.20 per second and Seedance 2.5 is $0.20 to $0.68. Per-clip and 100-clip table, with the Sume comparison.
Sneaker drop teaser in 6 seconds: Gemini Omni timecodes and cost
A 6-second vertical sneaker teaser built from a product photo and two timecoded beats in Gemini Omni on Sume, with 720p and 1080p prices and a Veo comparison.
Retry or fix? A status-code policy for Sume API clients in Python
Retry 429 and 503 with the same idempotency key; never retry 400, 401, 402 or 404. A small Python policy function, tested, with the Sume docs rules behind it.
Sume job webhooks: 10 attempts, 30 seconds apart, at least 270 s
Ten attempts with a fixed 30-second gap cover at least 270 seconds. What that means for your deploys, cold starts and when to fall back to polling a Sume job.
Voice note to SRT in Python with Sume STT sentence segments
Submit a voice note to Sume STT with sentence segmentation, poll the job, and write an SRT file from segments[]. Runnable Python with asyncio.run.
Suno v6 retires earlier models: pin an engine id for ad music
Suno says previous models retire as v6 rolls out. If an ad needs the same bed next quarter, pin an explicit Music Router id on Sume instead of auto.
AI talking avatar video cost per minute: Sume tiers
Sume avatar video is $0.184, $0.245 or $0.55 per second by tier: $11.04, $14.70 or $33.00 a minute. 100 thirty-second clips cost $552 to $1,650.
TikTok ad creatives prohibit watermarks: sample corner stills first
TikTok's reservation spec prohibits watermarks, TikTok's own included. Pull stills with video-inspect and crop the four corners before a reused clip goes up.
TikTok reservation ads need 2,500 kbps: compute it from the probe
TikTok's auction page asks for 516 kbps, its reservation and TopView pages for 2,500 kbps. Compute the average from video-inspect size and duration first.
TikTok TopView locks assets at preloading: freeze the cut first
TopView creatives need sales-rep pre-approval and cannot change once preloading begins. Use Timeline's unbilled plan, then render once with an idempotency key.
Cost of 100 AI avatar UGC ads: script to captions on Sume
100 fifteen-second avatar ads with captions cost $296.00 (Standard), $387.50 (Plus) or $845.00 (Max) on Sume, plus $0.95 once to create the avatar.
UGC-style ad without a creator: close-camera or usage demo?
Two Sume catalog Formats make UGC-style ads: sume-close-camera-ugc for a selfie testimonial and sume-product-usage-demo for hands-on use. How to choose one.