Tofu Models by Bland
1 model2026
Last refreshed 2026-11-14. Next refresh: weekly.
About
Tofu is Bland's line of full-duplex speech-to-speech models for enterprise phone calls. The first release, Tofu-1, was announced on November 13, 2026 as a research preview.
Decision facts
- Best fit
- audiorealtime voicevision and multimodal work
- Capability starting point
- Tofu-1 with multimodal inputs
- Lowest tracked input
- Not tracked
- Closest related family
- AssemblyAI
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
1 in view
Tofu-1Current
Use when the workload needs realtime voice, multimodal inputs, and audio.
2026-11realtime voicemultimodal inputsaudio
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Tofu-1 | Use when the workload needs realtime voice, multimodal inputs, and audio. | 2026-11 | realtime voicemultimodal inputsaudio | Current |
Release Timeline
1 release group2026-11
1 current
Tofu-1
Currentrealtime voicemultimodal inputsaudio
Specifications(1 models)
| Model | Released | Multimodal |
|---|---|---|
| Tofu-1 | 2026-11 | Yes |