Guides
Sume guides: the API, Formats, agents, models, media tools and pricing, each answer dated and sourced.
Wan 3.0 or Seedance 2.5: which one for which brief
Pick Wan 3.0 for cheap 2 to 30 s drafts and heavy reference sets; pick Seedance 2.5 for ByteDance's long-take behavior. A decision table from Sume's catalog.
Which Sume video models take 30 seconds: four ids in the catalog
Four ids in Sume's video catalog reach 30 seconds: seedance-2.5, wan-3.0, higgsfield-genjutsu and h3-max-recast. Which make a clip from a prompt.
Wan 3.0 vs Seedance 2.5 at 30 seconds: the price gap on Sume
A 30-second 720p clip costs $3.75 on Wan 3.0 and about $17.33 on Seedance 2.5 on Sume, a gap of 4.6x. A table across resolutions and what else differs.
First and last frame video on Sume: Seedance 2.5 or Wan 3.0
Both seedance-2.5 and wan-3.0 take a start image and an end image on Sume. How to send frame_images, and why you cannot mix them with references.
Shortest AI video on Sume: Wan 3.0 at 2 s, Seedance 2.5 at 4 s
Wan 3.0 takes 2-second clips on Sume; Seedance 2.5 starts at 4 s. A 2 s Wan clip at 480p is $0.125 on Sume. Duration floors across the catalog.
Reels on the Instagram TV app: caption for a living room
Reels on a TV autoplay with sound, but a screen across the room needs bigger text. Raise font_size_ratio in a Sume caption job and test one frame first.
Will my AI Reel show on Instagram for TV? 13+ ratings, shared screens
Instagram says TV Reels are for shared viewing and follow 13+ ratings. What that means for AI Reels, what the page leaves out, and a frame check before posting.
Instagram for TV timeline: Fire TV test to Google TV, what to do
Reels reached the living-room screen in a US test on Fire TV, then Google TV. The dated facts, what is unknown, and a short checklist for Reel makers.
Which AI video models take an audio file as a reference on Sume?
MiniMax H3 and H3 Max, Wan 3.0 and the Seedance 2.x rows accept audio references on Sume; Omni and Kling do not. H3 limits: 3 clips, 2-15 s each, 15 s total.
Check MiniMax H3 reference limits before you submit: a Python guard
Sume's minimax-h3 rows take up to 9 images, 3 videos and 3 audios, 12 in all, and audio cannot be the only reference. A 20-line Python guard catches it.
What an Omni video edit costs on Sume: the 8-second reserve hint
An Omni edit sends no duration to the provider, so Sume sizes the estimate from a hint (default 8 s). Table for 3, 5, 8 and 10 s at 720p, 1080p and 4K.
Omni in the Gemini app takes 5 photos; the Sume row takes 10 + 3 clips
Google's Omni page lists up to 5 photos per video. The gemini-omni-flash-1.1 row on Sume takes 10 images and 3 clips of 3 s, with <IMAGE_REF_0> tokens.
Omni replaces Veo in the Gemini app: does your API code change?
Google's page says Omni replaces Veo in the Gemini app. What that means for API code, and the one Omni row Sume lists, with its limits and per-second price.
MiniMax says H3 edits across modalities. Which Sume row edits a clip?
MiniMax lists reference and editing across modalities for H3. On Sume, H3 does text, image and reference video; clip edit by prompt is Omni's video_url mode.
Replace the background in an AI video: Ray3.2 or an Omni edit on Sume?
Luma's Ray3.2 lists background replacement with the original lighting kept. Sume's route is an Omni video_url edit: prompt, source clip, optional resolution.
MiniMax H3 open weights: what is released and what stays API-only
MiniMax opened H3-Base on 2026-08-03 with two checkpoints, but 2K regeneration and Context-IR stay API-only. How that maps to Sume's minimax-h3 rows.
First Draft only cuts clips you have: generate the b-roll first
A starting edit needs footage. Generate 9:16 cutaways with Seedance 2.5 or Gemini Omni Flash 1.1 first: durations, aspect ratios and one Video Router call.
Reel auto-captions misspell names and brands: fix with script_text
Auto captions stumble on names and brands. Send your script as script_text so Sume keeps the speech timings and prints your spelling. Errors and the $0.20 job.
Caption a Reel after First Draft: burn once on the final cut
First Draft shifts where every word lands, so burn captions only on the final cut and restyle with source_caption_id. The $0.20 job, errors and corrections.
Pause threshold test: 0.4 s versus 1 s silence splits on one clip
Before First Draft or your own trim removes pauses, test what a pause is: run video inspect with silence_split_seconds at 0.4, 0.7 and 1.0 on one clip.
Is there an API for Instagram First Draft? Sume job for each step
The pages we read describe First Draft as an in-app feature, not an API. Each action maps to a Sume job and price, plus the one step that stays manual.
First Draft, CapCut Auto Cut or Firefly Quick Cuts for Reels?
Three first-pass editors assemble clips into a starting cut. Where each runs, who it suits, and where an API cut fits when the footage is a folder, not a phone.
First Draft cuts one Reel at a time: rough-cut 20 clips by API
Instagram's First Draft runs inside the app on the clips you picked. For a batch of 20 talking clips, trim and join them by API: cost table, steps, one curl.
Seedance 2.5 with a reference video: how Sume prices a 30 s job
Sume's Seedance estimator adds 15 assumed input seconds when a reference video is present and multiplies by 0.6. At 720p, 10 s costs $8.67, not $5.78.
Seedance 2.5 or Seedance 2 on Sume: price and length compared
Seedance 2.5 runs 4 to 30 s at $0.58 per second at 720p on Sume; Seedance 2 caps at 15 s and lists at $0.014 per 1,000 tokens against $0.0214. A table.
callback_url for 30-second Seedance 2.5 and Wan 3.0 jobs
Add callback_url (HTTPS) to a 30-second seedance-2.5 or wan-3.0 job and Sume POSTs a signed webhook when it finishes. Headers and what to verify.
Retry a failed 30-second video job on Sume without paying twice
Reuse the same Idempotency-Key when you retry a seedance-2.5 or wan-3.0 submit on Sume; a replay returns the original job. Failed status and error field.
Seedance 2.5 price per second on Sume: 480p, 720p and 1080p
Seedance 2.5 on Sume costs about $0.27, $0.58 and $1.42 per second at 480p, 720p and 1080p in 16:9. How the token math and the 1.25 factor get there.
Image API stream:true returns 400: use async mode and events
Sume's Image API does not stream yet: stream:true returns 400 streaming_not_supported. Use mode async and read /events for progress, or a webhook.
Three 'auto' settings on GPT Image 2.5: aspect ratio, size and quality
auto means three different things on ChatGPT Image 2.5 through Sume: it matches an edit's shape, reserves an upper-bound size price, and reserves max quality.
GPT Image 2.5 4K returns 202: a Python poll loop that works
Slow ChatGPT Image 2.5 calls on Sume return 202 and a job envelope after 30 seconds. Python that handles 200 and 202 without paying twice.
Testing four new video models at once: how many jobs can Sume hold?
A model bake-off submits many jobs at once. Sume accepts 6 on Free, 24 on Pro, 48 on Startup, 120 on Scale, and queues the rest; past that is 429 queue_full.
Trailer Reel for your Instagram Series: 15 seconds from 3 episodes
Cut a 5-second beat from each of the first three episodes, join them with a short fade, and burn one title line. Steps, per-job prices and a curl.
Add old Reels to an Instagram Series: audit the back catalog first
Meta's Series test groups new and old Reels on one profile page. Before picking episodes, check each Reel's length, sound and opening frame with video inspect.
Pick episode one of your Instagram Series from Edits comparison
Compare up to three Reels in Edits, pick the one with the best watch time, then re-cut it as episode one with a clean opener, captions and a next-episode cue.
Native-speaker review for an AI voiceover: a 10-minute checklist
Decagon validates locale voices with native speakers. A checklist to do the same review on any AI voiceover before you publish it in another language.
Decagon Voice 3 handles 70+ languages mid-call. A TTS job takes one
Voice 3 detects language and switches mid-sentence. A Sume TTS job speaks one language per request. How to plan a multilingual script around that.
Silent AI video costs more: H3 always has sound, Kling 3 can drop it
MiniMax H3 always returns stereo audio, with no toggle on Sume. Kling 3 can switch audio off, and a silent 10 s at 720p still costs $1.40, above H3's $0.75.
OpenAI gave six months' notice on Sora: dates and a checklist
OpenAI told developers on 2026-03-24 that the Videos API and Sora 2 ids go on 2026-09-24, with no replacement listed. Six ids and a checklist.
Reference ingest for a 30-second ad: shots, text and audio in one read
Reference ingest reads one clip of up to 300 s into a manifest of shots, OCR text and audio facts. Use it before remaking a winning 30-second ad.
Wan 3.0 references on Sume: 10 images, 5 videos, 5 audios
Wan 3.0 takes more reference files per Sume job than Seedance 2.5: up to 10 images, 5 videos and 5 audios. How to send them and what the API rejects.
Seedance 2.5 references: ByteDance says 30 images, Sume allows 9
ByteDance describes up to 30 images, 10 video clips and 10 audio clips for Seedance 2.5. Sume's API request check caps one job at 9, 3 and 3, 12 files total.
4:5 still to video: no Sume video model lists 4:5, so pick another
No video model in Sume's catalog lists 4:5. Make the still 3:4 or 1:1, animate it, then crop to 4:5 with a $0.02 video-filter pass.
Gemini Omni Flash 1.1 from a still: 16:9 or 9:16 only, cost 360p to 4K
A still-to-video handoff to Gemini Omni Flash 1.1 on Sume must be 16:9 or 9:16. A 5-second clip costs $0.1875 at 360p up to $1.88 at 4K.
Edits on desktop or an API render: which for a weekly Reel?
Edits now has a desktop app and an AI assistant. Use it for creative one-offs and a Sume render for repeated, logged Reels: a decision table with job prices.
4:5 AI video: HappyHorse lists it, Sume's rows do not. Crop instead.
Alibaba's HappyHorse text-to-video reference lists 4:5, 5:4 and 9:21. None of Sume's H3, Kling or Omni rows takes 4:5, so generate 3:4 or 1:1 and crop.
HappyHorse result links expire in 24 hours: save the file the same day
Alibaba's HappyHorse API says video URLs and task ids last 24 hours. Download on completion, in Python, from Sume's content endpoint or any vendor's.
Wan 3.0 price per second: fal's list rate vs what Sume bills
fal lists Wan 3.0 at $0.05, $0.10 and $0.20 per second. Sume bills that list times 1.25: $0.0625, $0.125 and $0.25 per second at 480p, 720p and 1080p.