Models

Open-weight models for realtime voice and agent workloads. Click a card for specs, sources, and how to call the API.

Model APILLM

Gemma 4 31B IT

31B · Dense · Instruction-tuned

Google DeepMindView →
Model APILLM

Gemma 4 26B A4B IT

26B MoE · ~4B active · Instruction-tuned

Google DeepMindView →
Model APILLM

Qwen3.8-27B

27B · Dense · Vision-language

QwenView →
Model APILLM

DeepSeek V4.1 Flash

552B MoE · 8-16B active · Text and images

DeepSeekView →
Speech-to-text

Qwen3-ASR-1.7B

1.7B · Streaming speech-to-text

QwenView →
Model APIJev

Gemma 4 31B Jev

Calibrated answers · Text and images

Google DeepMindView →
Model APIJev

DeepSeek V4.1 Flash Jev

Calibrated answers · Text and images

DeepSeekView →