What resolution is Tavus Griffin? 720p, and Sume avatar clips
Tavus says Griffin generates 720p video in 320 ms chunks. Sume avatar video renders at 720p too, as a clip. What the same number means in each product.
Sume Avatar 1.0
Sume Avatar 1.0 by API: reusable avatars, talking video from a script, multi-scene and two-speaker videos, previews, B-roll, and Face Swap (Beta).
Start with: Introducing Sume Avatar 1.0
Viewers suspect AI within 20 seconds: disclose in the first frame
Tavus says Griffin-Lite doubters suspected within 20 seconds, and participants were told only afterward. Put the AI disclosure up front in a Sume clip.
Add a pause to an AI avatar video with a silence beat
To make a Sume avatar stop talking for a few seconds, use a silence scene in video_inputs: duration required, no script allowed. Example, limits and cost.
AI face swap video cost: the $2.76 to $8.25 ceiling per clip
Sume's Beta avatar face swap reserves a 15-second maximum: $2.76 on standard, $3.68 on plus, $8.25 on max. Constraints and when to use talking video instead.
Avatar 1.0 talking video: 4 to 60 seconds, cost by quality tier
Work out what a 4, 15, 30 or 60 second Avatar 1.0 talking video costs at standard, plus and max quality, using the per-second rates in the Sume catalog.
Avatar video preview: approve the first frame, then pick quality
Sume avatar video previews are tier-independent: approve stills, then render at standard, plus or max. What can change at generate-video, and what cannot.
Create an AI avatar from a reference image: URL rules and cost
Sume turns a public HTTPS photo into a reusable avatar for $0.95. The request, the URL checks that reject a bad image, and how to use the handle in videos.
Lost an avatar video job id? List avatar videos instead
Sume has GET /v1/avatar-videos and GET /v1/avatar-videos/:id. How to find a render after a crash without resubmitting, and which status field to read.
avatar-1.0/image-to-video is deprecated: switch to veed/fabric-1.0
Sume's avatar-1.0 image-to-video routes are deprecated aliases of VEED Fabric 1.0. Same body, new URL: audio_url, duration_seconds and one image source.
Avatar handle with a leading @: how Sume stores and reuses it
Sume accepts an avatar_handle with or without a leading @ and stores it without. Why a stable handle beats a generated id, and how to reuse it.
Avatar preview resource_status vs job_status: which field to poll
Avatar video previews return resource_status and job_status next to a legacy status field. Read resource_status for readiness and job_status for polling.
Cancel an avatar video job: the 409 job_generation_already_started
You can cancel an avatar video job only before generation starts. After that Sume returns 409 job_generation_already_started and the job runs to completion.
Create an AI avatar from profile traits: the props input
Avatar 1.0 can build a reusable avatar from structured traits, not a prompt or photo. The props input takes ethnicity, sex and age. When to use it.
Face swap video_url rejected: signed and private URLs explained
Sume face swap needs a public HTTPS video_url. Signed or private URLs, localhost and provider task URLs are rejected before generation. How to host the clip.
HeyGen Avatar 3.0 singing and 177 languages vs Sume Avatar 1.0
HeyGen Avatar 3.0 adds singing and 177+ languages. Sume Avatar 1.0 renders script-driven talking video, 4 to 60 seconds. What each one covers.
Regenerate avatar preview stills, or start a new preview?
Regenerate refreshes first-frame stills from the stored preview request. Changing script, avatar, scene or aspect ratio needs a new preview. The full rule.
Sume Avatar API: canonical routes vs the legacy model-run aliases
Which Avatar 1.0 endpoint should a new integration call? The canonical /v1/avatar-1.0 routes, with the legacy aliases kept for compatibility. All paths listed.
Tavus Memory Stores vs Sume: personalizing avatar video per person
Tavus PALs now keep persistent memory per participant. Sume avatar videos are one-shot renders, so personalization is in the script you send. Here is the split.
Avatar video soundtrack: send prompt or audio_url, volume 0.05-0.4
Sume's Avatar Video package accepts a soundtrack with exactly one of prompt or audio_url and a volume from 0.05 to 0.4, default 0.15. What each costs and does.
Dub with lip sync: Meta Reels option vs Sume Avatar 1.0 (English-only)
Meta offers optional lip sync on translated Reels. Sume Avatar 1.0 is English-only, so a non-English talking shot uses TTS plus a lip-sync endpoint.
Gemini Omni avatar with your own voice vs Sume Avatar 1.0 video
Google says Gemini Omni can make an avatar video in your own voice. Sume Avatar 1.0 takes a script, 4 to 60 seconds, 720p. What differs.
One avatar handle, three platform cuts: keep the character consistent
Keep one presenter across LinkedIn, Snapchat and Pinterest by reusing a single avatar_handle and rendering each cut at the right ratio and length with Sume.
Snapchat Spotlight needs 6 seconds: avatar clips that start at 4
Snap's Public Profile API takes Spotlight videos of 6 to 60 seconds, but Avatar 1.0 plans from 4. Plan scripts at 6 seconds or more and probe before posting.
Synthesia accepts WebP with transparency: what Sume avatar URLs need
Synthesia added WebP support on Sep 29, 2026, animated or static with transparency kept. Check what image URL rules apply when you feed Sume Avatar 1.0.
Tavus Griffin-Lite is not available: what to render today
Tavus says Griffin-Lite is a research preview for select testers only. If you need an avatar video this week, here is what Sume Avatar 1.0 renders today.
Tavus Griffin single photo input vs a Sume photo avatar handle
Griffin works from one reference photo. In Sume you turn one photo into a named avatar handle once, then reuse it in every script. Steps, inputs and cost.
Avatar catalog estimate is a 4-second floor: real cost at 15, 30, 60 s
The catalog shows 74, 98 and 220 cents for avatar videos, which is the 4-second floor. Here is what 15, 30 and 60 second scripts cost at each Sume quality tier.
Avatar video captions: inline has four knobs, no design override
Inline captions on a Sume avatar video take style, font, language and script_text. For design colors, caption the clean MP4 with standalone captions.
Sume avatar video stuck? Read the job events before retrying
Use GET /v1/jobs/{id}/events to see whether an avatar video is queued, started, submitted or failed before you cancel, wait, or resubmit and pay twice.
Avatar video scene_previews: one still per scene, one shared set
Multi-scene avatar previews return scene_previews, one still per scene. Later stills continue scene 0's pose, so review scene 0 first.
Avatar 1.0 media URLs: image_url, product_image, scene.image_url
Which Sume field takes which public HTTPS URL: input.image_url, product_image, scene.image_url, background.url, video_url. Table, rejection rules, examples.
Avatar video_inputs voice rules: text, silence, script vs input_text
Each video_inputs scene takes a voice of type text or silence. Text needs exactly one of script or input_text; silence needs duration and forbids both.
Can an AI avatar be your TikTok Shop affiliate creator?
No: TikTok Shop affiliate creators need 1,000 followers, 18+, and ID plus live-selfie verification. What an avatar can still do for a seller, and where.
Gemini app video avatar: the @username prompt vs a Sume avatar handle
In the Gemini app you add your avatar to a video by typing @ and your Google username, 18+ only. Sume uses an avatar_handle and a photo input. What differs.
Korean avatar video captions: fix caption_hangul_text_latin_style
A Korean script with captions style slam, punch or tiktok-green is rejected with 400 caption_hangul_text_latin_style. Use a Hangul style instead.
Find the avatar handle for sume avatar-videos create (--agent --json)
List avatars with sume avatars list --agent --json, pick the avatar_handle, then run sume avatar-videos create --confirm-paid. Flags and limits explained.
Cost of 10 eight-second avatar ad hooks: Standard, Plus, Max
Ten 8-second Sume Avatar 1.0 hooks cost $14.72 on Standard, $19.60 on Plus and $44.00 on Max at list rates. Plus is the default if you omit quality.
Avatar video progress bar: subscribe mode sends no progress events
Mode subscribe on an avatar job gets the same 30-second wait as sync, not a stream. Build progress UI from async status polling and job events.
Prime Video AI lip sync vs Sume: human voice on a talking image
Amazon uses AI and VFX to lip-sync shows to human-dubbed audio. Sume cannot re-sync existing footage, but Fabric can animate a still to a human recording.
Synthesia Roleplay Sessions vs Sume scripted scenario clips
Synthesia launched Roleplay and Survey Sessions on Oct 1. Sume has no live roleplay, but can render the scenario prompt as a short avatar clip.
Synthesia Survey Sessions vs Sume: not interactive, but scripted
Synthesia's Survey Sessions avatar asks questions live and reports themes. Sume renders avatar question clips from a script, with no live conversation.
VEED Fabric Emotions audio tags vs Sume Fabric audio_url
VEED's Fabric Emotions adds inline tags like [excited] in the script. On Sume, Fabric takes a finished audio_url; emotion lives in the TTS step.
HeyGen 30-minute avatar video: what Sume does in one request
HeyGen says it can make a 30-minute talking video in one pass. One Sume avatar video request covers 4-60 seconds, so longer pieces are split into jobs.
HeyGen Edit Look: retouching an AI avatar, and the Sume route
HeyGen's Edit Look retouches an existing avatar in place. Sume's docs show no such edit, so the route is a new avatar from a retouched photo and a new handle.
Does a face-swapped video need a YouTube AI disclosure?
YouTube asks for disclosure when content makes a real person appear to say or do something they didn't. What that means for a Sume Beta face-swap output.
Shortest video an AI dub or face swap accepts: 5 s vs 4 s
Synthesia's dubbing page says a video must be at least 5 seconds. Sume's Beta face swap plans for about 4-15 seconds and avatar videos take 4-60. Side by side.
Face swap or avatar talking video: which Sume endpoint?
Use face swap when you already have a 4-15 second source video; use the avatar talking video when you have a script. Inputs, limits and what each returns.
Two-host AI video podcast: what Sume avatar videos can do
HeyGen's Video Podcast puts two AI hosts in one studio scene. A Sume avatar video resolves one avatar per final video, so a two-host show is two renders.