Akool streaming avatar sessions: 15 to 60 minutes vs Sume clips

Akool lists a streaming avatar session of 15 minutes on Starter and Pro, 30 on Pro Max and 60 on Business, with 5 free minutes on its Free plan. Sume has no streaming session. It renders a script into a clip of 4 to 60 seconds that you can play any number of times.

Live session or recorded clip

A streaming session is for a conversation: the viewer talks, the avatar answers in the moment. A rendered clip is for a message you can write in advance, such as a greeting, an explainer or an ad.

What a Sume clip costs instead

If the viewer experience is a short, scripted answer, the cost is per second of video, and the audience can be any size. A 30-second clip is $5.52 on standard, $7.35 on plus and $16.50 on max at the catalog rates.

A full 60-second job is $11.04, $14.70 or $33.00. Anything longer is several jobs. Sume's avatar video guide is explicit that the window is 4 to 60 seconds.

Sizing a pre-rendered set

Suppose a help page has 12 likely questions, each answered in 20 seconds. That is 240 seconds, or $44.16 on standard, $58.80 on plus, or $132.00 on max. You pay once, and every visitor plays the same files.

A live session instead bills as a plan feature and ends at its cap, 15 minutes on Starter. After that, the viewer has to start another. A rendered clip never expires mid-sentence, because its length is fixed when you submit the job.

Where Sume does not fit

Sume does not hold a two-way conversation, listen to the viewer or answer a question it has not seen. If the product is an agent that talks back, use a live avatar product. A common middle path is to pre-render the 10 to 20 answers a visitor asks most, and route the rest to a human or a chat.

A decision rule

Ask whether the viewer needs to say something back. If yes, you need a live product, and Akool's streaming cap becomes a design constraint: plan for a session that ends at 15 minutes, 30 or 60 depending on plan. If no, render a clip and spend your time on the script instead.

For many sites the right answer is both: a rendered welcome clip at the top of the page, and a chat or human handoff below it.