Griffin-Lite waitlist: six checks before planning a live avatar launch
Tavus Griffin-Lite is a research preview with no customer date or price on its page. Six checks before you plan around it, and what to ship on Sume meanwhile.
If you are planning a launch around Tavus Griffin-Lite, six facts decide whether the plan is safe: whether you can get access, what it costs, what the disclosure rules will be, how many sessions run at once, what happens when it fails, and what you ship in the meantime. As of 2026-10-09 the Tavus Griffin page answers the first two with 'select early testers' and 'no price listed'.
The six checks
Each row is a question to put to the vendor in writing. The right-hand column says what the Tavus page currently shows, so you can see which questions are open.
| Question | What the page shows |
|---|---|
| Can customers use it today? | No: 'will not be available for use for customers at this time' |
| What does it cost? | No price on the Griffin page |
| What is the disclosure policy? | Page says safe disclosure features are being worked on |
| How many concurrent sessions? | Not stated for Griffin-Lite |
| What is the failure behaviour? | Not stated |
| What is the access route? | Request form for the waitlist |
What the existing Tavus plans do tell you
Tavus's pricing page does list limits for its current developer plans: Basic has 1 concurrent stream and 25 minutes, Starter 3 concurrent streams with 100 minutes at $59 a month, and Growth 10 concurrent streams with 1,250 minutes at $397 a month. Those are not Griffin-Lite terms, and the page does not say that Griffin-Lite will inherit them. Use them to size an interim pilot, not as a promise for the preview.
What to ship on Sume while you wait
If part of your plan is a scripted message, you do not need to wait. Sume's avatar video route renders a known script with a ready avatar: $0.184 per second at standard, $0.245 at plus, $0.55 at max, within a 4-60 second window (docs and catalog, read 2026-10-09). A 30-second welcome is $5.52 at standard. Create the avatar once ($0.95), approve a first frame with a preview, then render.
Keep the avatar handle so later clips reuse the same face. What you should not do is promise customers a live agent on a date the vendor has not given you.
- Pilot the recorded path now.
- Record the questions above in writing with Tavus.
- Revisit when the page lists customer availability.
Writing the plan down
Put the answers in a one-page plan with three columns: known, unknown, and fallback. Known: Griffin-Lite is a preview for select testers, the page lists latency figures and a 48 percent result from a one-minute study of 54 participants, and says disclosure features are in progress. Unknown: price, date, concurrency, failure behavior. Fallback: a recorded avatar clip rendered on Sume, with a clear label that it is a recording.
Review the plan when Tavus updates its page. A change from 'not available' to a price and date is the signal to re-plan; a press article is not. Tavus's own Griffin page is the source to check.
Sources
Related posts
More in Use cases
- Headline card above an app screen recording: compose stack, $0.36
Timeline compose stack puts a headline image over a video for $0.02 per job. Three headline variants over one app recording cost $0.36 with the images.
- Hide jump cuts with alternating punch-ins: one crop copy, N slots
One $0.02 video-filter crop makes a 1.25x punched-in copy. Alternate wide and punched slots at each pause cut for $0.02 total, not $0.02 per cut.
- How many words fit in a 30-second voiceover, and what it costs
A 30-second voiceover is about 75 words or 450 characters at a planning pace of 150 wpm. Sume TTS bills characters, so here is the length-to-cost table.
- iMessage app has no App Store preview: where the video goes
Apple's specification page says app previews are not available for iMessage apps. Keep the demo clip for your site and social, made with the Sume Format API.
Written by Sume