One narrator across a series: Sume TTS 1.0 is Sonic 3.6, pin older ids
Sume TTS 1.0 always runs on sonic-3.6. To keep a series that began on an older Sonic, use the TTS Router and pin sonic-3.5 or sonic-3 with the same voice.
POST /v1/tts-1.0/generate has no engine picker and always runs on sonic-3.6, the current stable Sonic. If a series began on an older engine and you want episode twelve to sound like episode one, call POST /v1/tts-router/generate with a required model such as sonic-3.5 or sonic-3 and the same voice. Sending model to TTS 1.0 is a 400.
The catalog is read from the Sume TTS router API (GET /v1/tts-router/models) described in the Sume API reference (read 2026-10-06).
Which ids can I pin?
The router's v1 catalog has five ids, all Cartesia Sonic.
| Id | What it is |
|---|---|
sonic-3.6 | Current stable Sonic; what TTS 1.0 uses |
sonic-3.5 | Earlier Sonic release |
sonic-3 | Earlier Sonic release |
sonic-latest | Alias that resolves to sonic-3.6, never the preview |
sonic-preview | Beta channel; can change without notice and rejects pro voice clones |
How do I keep the voice the same?
The router uses the same voice selectors as TTS 1.0: avatar_id, avatar_handle or voice.id. It does not make a second voice namespace. Pin an exact id in the series config, not sonic-latest, so a future alias move does not change the sound mid-season.
What should I test?
Listen to the same paragraph on sonic-3.5 and sonic-3.6 with your voice before you commit the series. The job.model field on each result records the id you asked for, so you can audit which engine voiced each episode.
Sources
More in Models
- Kling 3.0 15-second clip: $2.10 silent or $3.15 with sound on Sume
Kling 3 on Sume makes 4 to 15 second clips at 720p or 1080p, with audio priced at $0.21 a second and silence at $0.14. Limits, ratios and how to try it.
- Kling 3 aspect ratios on Sume: 16:9, 9:16 and 1:1 only
kling-3 on Sume accepts 16:9, 9:16 and 1:1. For 4:3, 3:4 or 21:9 pick Seedance 2.5, Wan 3.0 or MiniMax H3 instead. A 400 means the value is not on the list.
- Kling 3.0 on Sume: kling-3 is v3 Pro, motion control is separate
On Sume, kling-3 is Kling Video v3 Pro for prompt and frame video; Kling 3.0 Motion Control is a different surface billed on the driving video's seconds.
- Longest single AI video clip by API in 2026: 30 seconds, two Sume ids
Seedance 2.5 and Wan 3.0 make a 30-second clip in one job on Sume. Kling 3 and MiniMax H3 stop at 15 s and Omni at 10 s. Limits and per-second prices.
Written by Sume