Blog
Sume blog: agent-native video generation, Avatar 1.0 orchestration, and production APIs for builders and coding agents.
Convert an SRT file to Sume caption cues in Python
Sume captions take no SRT upload, but cues carry the same start, end and text. A 26-line Python script turns an SRT into cues and posts them for $0.20.
Customer onboarding email series with an avatar: 5 clips, cost by tier
Five 20-second onboarding clips from one Sume avatar cost $18.40 on standard, $24.50 on plus and $55.00 on max. How to plan the series and when to re-render.
Do I pay for a failed AI avatar video job? Refunds on Sume
Sume reserves the avatar video price at submit, captures it on completion, and releases or refunds it where a job fails. What it means for retries.
duration vs duration_seconds on each Sume video route
/v1/videos takes duration; motion control and lip-sync take duration_seconds; recast and edit read the source clip. One table of what each does.
What does a first-time home buyer video series cost with an avatar?
Six 30-second avatar tips cost $33.12 at standard, $44.10 at plus or $99.00 at max on Sume, plus $0.95 once for the avatar. Rates and script limits.
Fix an underexposed photo: curves first, AI edit only if needed
Underexposed photo? Try Pillow autocontrast and gamma for free, then an ideogram/ideogram-v4.5 edit at $0.075 only if noise or colour needs more.
FLUX 3 Image is on BFL's pricing page: which FLUX ids can Sume call?
bfl.ai/pricing lists FLUX 3 Image beside FLUX.2 pro, flex, max and klein. On Sume you can call FLUX.2 Pro and Flex today. A dated table and the migration rule.
How do I turn my free course lessons into text for a sales page?
Detach each lesson's audio ($0.01), then transcribe it at $0.01 a minute: $0.09 for an 8-minute lesson, $0.90 for ten, on Sume.
Character still first, then Kling 3.0 motion control
Kling 3.0 motion control on Sume needs a still and a motion video. Make and approve the still first, then spend the $0.1575-a-second job on it.
GPT Image 2.5 quality ladder: price multiplier from low to max on Sume
At 1024x1024 on Sume, GPT Image 2.5 high costs 8.9 times low, xhigh 15.9 times and max 35.7 times. The ladder with billed prices and batch costs.
How do I make a Halloween party invitation video with RSVP?
Make the poster with Ideogram 4.5 ($0.075), animate it for 6 seconds ($0.75) and burn the date and RSVP as captions ($0.20): $1.025 on Sume.
How do I make a 9-minute Halloween haunt sound loop from one track?
Generate one 3-minute ambience track ($0.125) and join it three times with Timeline audio concat ($0.01): a 9-minute file for $0.135 on Sume.
Handle every Sume API error with one switch on next_action
Sume errors share one envelope. Branch on next_action, retryable and retry_after_seconds, and your client handles new codes without a code change. JS sample.
Hindi speech to text API: Sume STT with language_code hi
Transcribe Hindi audio with Sume STT: send language_code hi, check the reported language, and review code-mixed speech. $0.01 per audio minute.
How much do six gift guide cover images cost at each quality tier?
Six Ideogram 4.5 cover options on Sume: $0.225 at low, $0.45 at medium, $1.65 at high. Which tier to use for which step.
How much narration per product in a gift guide video, and its cost?
Plan about 150 characters a product plus 300 for intro and outro. Eight products is 1,500 characters, $0.08 on Sume TTS, or $0.09 as eight jobs and a join.
How many AI music tracks should I generate for one ad before picking?
Each Music Router track is a flat $0.125, so three takes cost $0.375 and eight cost $1.00. Here is a take budget by how fixed your brief is.
How to shoot 1 to 4 person photos for H3 Max Recast
Photo guidance for Recast references on Sume: one photo per new person, framing, light and angle. What the route takes and what to check before you submit.
Image-to-video not starting on my photo: frame_images vs references
Your photo is a reference, not a first frame, when it goes in input_references. Use frame_images with first_frame on Sume /v1/videos to pin the opening shot.
Image-to-video prompt: write the motion, not the photo
When a photo is already pinned as the first frame, spend the prompt on motion, camera and sound, not on describing what the picture shows. A worked rewrite.
Is GPT Image 2.5 cheaper than Seedream 5.0 Lite? It depends on quality
On Sume a 1024x1024 ChatGPT Image 2.5 costs $0.0074 at low, $0.0165 at medium and $0.0659 at high; Seedream 5.0 Lite is a flat $0.04375. Where they cross.
Japanese speech to text API: Sume STT with language_code ja
Transcribe Japanese audio with Sume STT: send language_code ja, read word times, and test a sample first. $0.01 per audio minute, 10 minute jobs.
Job id or run id? Which Sume endpoint to poll for each product
Jobs, Format runs, Actions and Agent Completions have different ids, poll URLs and webhook events. Which to poll for each product, and which SDK helper to call.
Landscape listing photos in a vertical reel: cover, contain or blur?
A 3:2 photo in a 9:16 reel shows only 37.5% of its width with fit cover. Pick the fit per photo in Timeline 1.0: eight photos and a bed cost $0.225 on Sume.
Let an agent pick talking video, Fabric or H3 Max over MCP
Rules for an agent on hosted MCP: avatar-videos_create for scripts, avatar-image-to-video_create for audio, and dry_run plus max_spend_usd before any paid call.
Let browsers start Sume jobs through your server, not with your key
Browsers must never hold a Sume API key. A server route authenticates the user, checks the input, derives an Idempotency-Key and returns only the status URL.
Lip sync 30 seconds as three 10-second lines: cost by resolution
MiniMax H3 Max lip sync caps audio at 14.8 s, so a 30-second script becomes three jobs. Cost at 480p, 768p and 1080p, with ceil rounding and a split rule.
Lock the look with one image, then animate it on three video models
Make one approved image, then use it as first_frame on wan-3.0, kling-3 and minimax-h3-max. Same start, three motions, one prompt, so you can compare.
Logo screenshot to a transparent PNG: RMBG or GPT Image 2.5 background
Make a transparent logo PNG from a white-background screenshot: Sume RMBG ($0.0225) keeps your pixels; GPT Image 2.5 background transparent redraws the logo.
Mandarin Chinese speech to text API: Sume STT language_code zh
Transcribe Mandarin audio with Sume STT using language_code zh, then check the result and timings. $0.01 per audio minute and no accuracy claim without a test.
Meta Feed allows 1-second video ads: shortest Sume clip is 2 s
Meta's Feed spec allows video from 1 second. The shortest clip any Sume video model generates is 2 seconds (Wan 3.0). What to do for a 1-second cut test.
Migrate a real-time avatar prototype to Sume async jobs: what changes
Moving from a live avatar session to Sume means replacing a stream with submit, poll and fetch. The code changes, the UX changes, and a Node example that runs.
Nano Banana 2 is retired on Sume: same price on Nano Banana 2.1?
Sume retired google/nano-banana-2 and runs old requests as Nano Banana 2.1 at the same price card: $0.10 at 1K billed. Tiers, ids and what the job stores.
How do I make one intro bumper and reuse it in every course lesson?
Make a 5-second bumper once ($0.83 with a sting), then add it to 24 eight-minute lessons at $0.82 each: $20.51 for the course on Sume.
Pick clip lengths from a voice-over script: one sentence per clip
Turn a voice-over into clips: measure each voiced sentence, round up, and pick the models whose duration window contains it. Windows for six Sume models.
Pin the model id in an ad test: sume/auto follows the catalog
sume/auto is a pure function of the request plus the catalog version, so two ad arms made weeks apart can land on different models. Pin an explicit id in tests.
Pre-rendered avatar greetings per visitor segment, not a live avatar
Instead of a live avatar for each visitor, render one short Sume avatar clip per segment ahead of time. Per-tier cost for six 12-second greetings.
What prompt makes a 15-second Black Friday ad track?
Write the track as three timed sections: build, drop, sting. The Music Router takes it for a flat $0.125; the whole 15-second ad is $2.305 on Sume.
Python: cheapest Sume image model that lists your aspect ratio
A 25-line Python script reads Sume's image catalog, keeps models that list your aspect ratio, prices each from its endpoints record and prints the cheapest.
What is a Q4 video budget for a one-person shop, week by week?
Ten weekly jobs from Halloween to New Year, built from 720p Omni clips, Ideogram images, voice, music and captions, come to $9.675 on Sume.
Recraft V4 or FLUX.2 Pro for an illustration set on Sume: price, edits
Recraft V4 bills $0.05 per image on Sume and is text-only; FLUX.2 Pro bills $0.0375 and takes reference images. Ratios, n limits and which fits a set.
Remove dust, lint and wrinkles from a product photo with an AI edit
Clean dust, lint and fabric wrinkles from a product photo with ideogram/ideogram-v4.5: one reference, no mask, $0.0375 to $0.275 per image on Sume.
Remove the photographer's reflection from a window photo: masked edit
Remove your own reflection from a shop window photo: mask it and send a gpt-image-2.5 edit with mask_url, $0.0094 to $0.0835 per image on Sume. Checks inside.
Replace an old logo on 40 product photos with Ideogram 4.5
Rebrand 40 product photos: send each photo plus the new logo as a second reference to ideogram/ideogram-v4.5. Cost on Sume: $1.50 at low, $3 at medium.
Replace one prop in a video with AI: bottle to apple edit prompt
Swap one object in a finished clip with gemini-omni-flash-1.1 on Sume: the docs example prompt, fields you cannot set, and an 8-second price.
Rotate the Sume webhook signing secret without dropping a delivery
Upgrade the verifier first, rotate with POST /v1/webhooks/signing-secret/rotate, deploy the new secret inside the 24-hour two-signature window, then confirm it.
Runway Dev image_to_video, video_to_video: one Sume route?
Runway Dev names image_to_video, video_to_video and text_to_image paths. Sume's POST /v1/videos infers the mode from your fields. What it changes in ad code.
Seedream 5.0 Lite or Nano Banana 2.1 for product shots on Sume
Seedream 5.0 Lite bills $0.04375 per image on Sume against $0.10 for Nano Banana 2.1 at 1K. Ratios, tiers, references and when the extra cost is worth paying.