TTS Router ids: sonic-3.6 to sonic-preview, one price book
Sume TTS Router lists sonic-3.6, sonic-3.5, sonic-3, sonic-latest and sonic-preview, all at the TTS 1.0 character price. What each id is and when to pin it.

The TTS Router accepts five model ids: sonic-3.6, sonic-3.5, sonic-3, sonic-latest and sonic-preview. Every one bills from the same character book as TTS 1.0, $0.0475 per 1,000 characters at list times 1.25, with a 20,000-character maximum. Pinning a model changes the voice engine, not the price.
Unlike Music, the TTS Router has no automatic mode: model is required on POST /v1/tts-router/generate. List the catalog with GET /v1/tts-router/models.
The ids
| Id | Note |
|---|---|
sonic-3.6 | Current stable release. |
sonic-3.5 | Previous release, pinnable. |
sonic-3 | Earlier release, pinnable. |
sonic-latest | Alias for sonic-3.6; never sonic-preview and not a dated snapshot. |
sonic-preview | Provider beta channel; output and availability can change without notice. Not compatible with pro voice clones: those requests fail with voice_model_mismatch. |
When to pin
Pin a numbered id for anything you will re-render later and need to sound the same, such as a series with a recurring narrator. Use sonic-latest when you want improvements to arrive on their own, and accept that a re-render can differ from the first. Keep sonic-preview out of production jobs.
curl -X POST https://api.sume.com/v1/tts-router/generate \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: tts-router-001" \
-d '{
"model": "sonic-3.5",
"transcript": "Welcome back. Today: three ways to cut a long edit.",
"voice": { "id": "'"$VOICE_ID"'" }
}'Cost check
That 52-character line bills $0.01, because each job rounds up to a whole cent. A thousand such lines sent as separate jobs are $10.00. The same 52,000 characters sent in 20,000-character requests would be $2.47, so for bulk work, batch lines into larger scripts and cut the audio afterwards.
Related posts
More in Models
- 12 Sume video ids: which need a source video, which take text
Nine of Sume's twelve video ids take a text prompt. grok-imagine-video-1.5 needs an image; higgsfield-genjutsu and h3-max-recast need a video. Ranges per id.
- Veo 3.1 clip older than 2 days: it cannot be extended
Veo 3.1 extends only Veo-made clips kept 2 days. If yours expired, here is what Google's page allows and what Sume's Omni edit and new generation do.
- When Seedance 2 Mini or Fast is enough, and when to pay for 2.5
A decision guide for Seedance tiers on Sume: what Mini, Fast and 2.5 cost for the same clip, what only 2.5 can do, and which jobs belong on which tier.
- Which video model for a 12 s 9:16 clip with end frame and sound?
Seedance 2.5, Wan 3.0, Kling 3 and MiniMax H3 all take a 12-second 9:16 clip with an end frame on Sume. Prices run from $0.90 to $6.94; here is the pick.
Written by Sume