Blog
Sume blog: agent-native video generation, Avatar 1.0 orchestration, and production APIs for builders and coding agents.
Meta's video from static assets: what to prepare in Sume first
Meta made Advantage+ video generation from static assets generally available on Oct 6. What to generate in Sume first, and what stays inside Meta's tools.
Codex account-scoped MCP grant cleanup and Sume's 1-hour token
Codex 0.161.0 adds account-scoped grant cleanup for enterprise MCP authentication. How that sits with Sume's one-hour OAuth token and its revoke endpoint.
Course sale avatar clip in five aspect ratios: five Sume calls
One course-sale script, five aspect ratios: avatar talking-video takes 1:1, 3:4, 9:16, 4:3 and 16:9 per call at 720p, 4 to 60 seconds. Plan five calls, not one.
Node 26.11 isValidHeaderValue: check a Sume Idempotency-Key first
Node 26.11 adds http.isValidHeaderValue. Use it to reject a bad Idempotency-Key before POST /v1/images, with a fallback for older Node, and see its limits.
Allergen line on a holiday menu video: burn it in with caption cues
Add a fixed allergen and price line to a silent holiday menu clip with video-captions cues. No speech-to-text runs, and Sume does not check the allergen text.
AbortSignal.any: stop a Sume request on SIGTERM or after 5 seconds
Combine a shutdown signal and AbortSignal.timeout with AbortSignal.any, then tell a timeout from a SIGTERM stop on a Sume GET /v1/me call in Node.
LinkedIn video ad length: trim a 90-second cut to 30 s with Sume
LinkedIn allows 3 seconds to 30 minutes and recommends 15 to 30 seconds. Trim a longer cut with Sume video trim at $0.02 a job, then check size before upload.
Amazon Sponsored Brands video: cut a 20-second window with Sume trim
Amazon lists 6 to 45 seconds for Sponsored Brands video and prefers 20 or less. Use Sume video trim to cut an exact range from a longer clip for $0.02 a job.
Vidu Q4 Preview has a seed field; Sume video models reject seed
Vidu Q4 Preview image-to-video takes a custom seed. Sume video models report seed false and return 400 unsupported_parameter if you send one.
Veo 3.1 object insert and remove vs Sume's Omni video edit
Google lists object insert and removal in Veo 3.1. Sume has no Veo row; its edit mode is Gemini Omni Flash 1.1 with video_url on Video Router.
Synthesia dubbing at 80 credits a minute vs a Sume lip-sync chain
Synthesia lists dubbing with lip-sync at 80 credits a minute. A 10-minute dub on Sume chains audio detach, TTS and lip-sync segments. Costs by step.
Vidu Q4 Preview renders 4K for 16 s; on Sume only Omni lists 4K
Vidu Q4 Preview lists 540p to 4K with 10-bit colour. On Sume only Gemini Omni Flash 1.1 lists 4K, for 3 to 10 s at $0.375 a second.
Runway Characters is real-time; Sume Avatar video is a 4-60 s job
Runway Characters makes a real-time conversational video agent from one image. Sume Avatar 1.0 renders a scripted talking video of 4 to 60 seconds.
Sume webhook signature fails: check the secret fingerprint first
When a Sume webhook signature will not verify, compare the 12-character secret fingerprint header before you debug the HMAC. Includes a Python verifier.
Airtable keeps run history 2 weeks on Free: store your Sume receipts
Airtable run history lasts 2 weeks on Free, 3 years on Enterprise. Save the Sume run id and output URL in the record so a holiday audit does not depend on it.
Seedream 5.0 Lite 'online search': not a Sume image parameter
ByteDance says Seedream 5.0 Lite can use online search. Sume's seedream-5-lite row has no search field. What the row offers: ratios, 10 references, $0.04375.
Airtable Free has no Run a script action: start Sume runs elsewhere
Airtable says Run a script is not available on Free. Start Sume Format runs from a service you control, and let Airtable receive results as plain records.
Adobe's Cyber 5 forecast: plan five daily Sume bulk queues
Adobe projects $47.5B for Cyber 5, 17.3% of holiday online sales. Map the five days to five Sume bulk queues with one idempotency key each and a cap per item.
Vidu Q4 audio on or off vs Sume generate_audio by video row
Vidu Q4 Preview makes sound with the picture. On Sume, Kling 3 charges more with audio on; Omni and H3 always make audio, and Omni rejects generate_audio false.
Hedra Omnia: one image plus audio, and Sume's lip-sync route
Hedra lists Omnia as full-body animation from one image and audio. Sume's H3 Max lip-sync route also takes a still and audio. Inputs and price compared.
Synthesia Pro's 6,000 to 48,000 credits in minutes of dubbing
Synthesia's Oct 1 plans list Pro at 6,000 credits and 12,000 to 48,000 on 2x to 8x tiers. At listed rates that is 75 to 600 minutes of lip-sync dubbing.
Zapier Catch Hook URL is the same in test and live: test Sume safely
Zapier's Catch Hook URL does not change between test and live. Use Sume's webhook.test delivery to check the Zap before a real run sends a real result into it.
Threads API video max is 300 s: split a 21-minute talk into parts
The Threads API page lists video up to 300 seconds and 1 GB. Split a 21-minute recording into five trim requests with a small Python script. $0.02 per trim.
Five 20-word avatar sentences make five 8-second clips, not three
Why five 20-word sentences plan as five 8-second clips (40 s) in Avatar 1.0, and how pairing shorter sentences saves clips.
Make takes 300 webhook requests per 10 seconds: a Sume queue fits
Make accepts 300 incoming requests per 10 seconds. A 100-item Sume queue with per-item webhooks and 16 in flight stays far below that. When 429 and 400 matter.
Fliki API needs Premium; Sume lists API rate limits on Free too
Fliki says API access is on Premium and Enterprise only. Sume's docs list API key rate budgets from Free through Scale. What that means for cost.
Synthesia's 500 signup credits: 5 minutes of video or 1 survey reply
Synthesia lists 500 signup credits: about 5 minutes of video, 6 of lip-sync dubbing or 1 survey response at its listed rates. The same time on Sume in dollars.
MiniMax H3 defaults to 2K at MiniMax; on Sume the default is 768p
MiniMax's H3 page says 2K is the default. Sume's minimax-h3 runs native 480p or 768p with a 768p default; 2K and 4K are billed upscales.
Product photo to holiday video ad: attach the photo to a Format run
Turn a product photo into a video ad with the sume-product-commercial Format. Attach up to 30 images by URL or asset id; media URLs in input share that budget.
Real estate listing voice-overs: 25 homes at 700 characters
25 listing narrations of 700 characters cost $0.385 on MAI-Voice-2.1, $0.263 on Flash, and $1.00 on Sume TTS as 25 jobs. See the batched figure.
30-second video ad on Sume: one clip or two joined, priced
One 30 s Seedance 2.5 clip is $17.33 at 720p; two joined 15 s Seedance 2 Fast clips are $9.17 with the $0.10 Timeline join. Wan 3.0 30 s is $3.75.
4K 3840x2160 for YouTube: Sume Timeline output stops at 2160 per edge
YouTube lists 35-45 Mbps for 4K. Sume Timeline allows even width and height from 256 to 2160, so 3840x2160 is refused. What you can render instead.
720x1280 vs 1080x1920 for a 60 s Short: 40 MB vs 63 MB
YouTube lists 5 Mbps for 720p and 8 Mbps for 1080p. A 60 second Short is 40.38 MB vs 62.88 MB with audio. Set output.width and height in Sume Timeline.
A 12-minute video job costs 24 status reads at 30 s and 360 at 2 s
Poll cadence arithmetic for Sume video jobs: 30-second polls, the 2-second SDK floor, and the 20-minute client deadline, with a runnable calculation.
AI avatar disclosure line as the first caption cue on a Sume clip
Tavus says disclosure features are still being worked on for realistic avatars. On a rendered Sume clip you can author the label yourself with caption cues.
Bash script to submit, poll, and download a Sume video
A 22-line bash and jq script for POST /v1/videos: idempotency key, 30-second polls, a 20-minute cap, and exit codes 0 to 3 that a scheduler can read.
Budget 250 product shots: read each Sume model's price in code
A short Python script reads the pricing line from Sume's per-model endpoints route and prints the cost of 250 product shots. Reference rules and a worked table.
Bulk run item completed is not success: check before you publish
A Sume Format bulk child can be completed without being a success. Read each result and use per-child webhooks; no queue webhook exists.
Cancel queued video jobs when the offer changes: Pro plan example
On Pro, 4 jobs process and 20 wait in the queue (24 accepted). Only queued jobs can be canceled; a started job returns 409 job_generation_already_started.
Check a new Sume video model against your clip spec before you pay
Four catalog fields decide if a new video model fits your pipeline: durations, resolutions, ratios, and references. A Python check that submits no job.
Chinese and English text in images: Hy Image 3.5's claim, Sume's rows
OpenRouter says Hy Image 3.5 Preview renders Chinese and English text. Sume does not list it. How to test Chinese signage on the Sume rows that are listed.
Clef context: 65,536 on Cloudflare, 16,384 on the model card?
Cloudflare's Workers AI page lists 65,536 tokens of context for Clef; its Hugging Face card says 16,384 by default. Budget from the smaller. Cost math inside.
Cost of 200 five-second clips a month: hosted Sume ids vs a rented GPU
At Sume prices, 200 five-second clips cost $75 to $200 a month depending on id and resolution. The same sum for a rented GPU, with the break-even hours.
One minute of AI video: Runway Gen-4.5, Luma Ray3.2, Wan 3.0, Seedance
One minute of AI video costs $3.60 to $85.29 depending on the model: Runway $7.20, Luma Ray3.2 $3.60 to $14.40, Wan 3.0 $7.50 on Sume. Arithmetic inside.
Cost per viewer of a 20 second avatar clip at 10, 100 and 1,000 views
A rendered Sume avatar clip costs the same however many watch: $3.68 on standard. That is $0.368 per viewer at 10, $0.0368 at 100, $0.00368 at 1,000.
EU, UK or South Korea team: which open video weights you may use
HunyuanVideo and MiniMax H3 licenses exclude the EU, UK and South Korea. Wan 2.2 is Apache 2.0 and LTX-2.5 is worldwide with a revenue line.
Game NPC barks: 800 lines at 60 characters, voice cost
800 NPC lines of 60 characters are 48,000 characters: $0.72 on MAI-Voice-2.1-Flash, $1.056 on MAI-Voice-2.1, $8.00 on Sume as 800 jobs.
gemini-nano-banana-2.1 or google/nano-banana-2.1: which id Sume takes
Google's pricing page names the model gemini-nano-banana-2.1. Sume's image API takes google/nano-banana-2.1 and runs the old nano-banana-2 as 2.1.