Support matrix
This matrix follows the adapters and capability allowlists in the current code. “Supported” means integrated; it does not certify every model, device and platform combination through real-device acceptance.
Voice pipelines
| Capability | Integrated | Cascaded |
|---|---|---|
| Processing | One realtime model handles speech directly | Speech recognition → language model → speech synthesis |
| Default combination | Qwen qwen-audio-3.0-realtime-plus |
Volcengine ASR → DeepSeek deepseek-flash → Volcengine TTS |
| Speech interruption | Integrated | Integrated |
| Coding, search and memory tools | Available according to enabled capabilities | Available according to enabled capabilities |
| Camera frames in conversation | Not supported | Listed vision models only; off by default |
| Independent Loop Camera monitoring | Supported with a separate vision model | Supported with a separate vision model |
| iPhone text input | No editable input | Available when the host supports editable input |
Integrated mode needs fewer settings. Use cascaded mode to change the language model or enable conversation vision. Both require credentials for the selected services.
Integrated mode can also use OpenAI (INTEGRATED_PROVIDER=openai, default gpt-realtime-2.1-mini, OPENAI_API_KEY) or Gemini Live (INTEGRATED_PROVIDER=gemini, default gemini-3.8-live, GEMINI_API_KEY); see OpenAI and Gemini providers. StepFun (INTEGRATED_PROVIDER=stepfun, preview) uses STEPFUN_API_KEY, and the Qwen model can be switched to qwen-audio-3.1-realtime-plus (QWEN_REALTIME_MODEL).
Cascaded language models
| Provider | Default model | Conversation vision | Credential |
|---|---|---|---|
| DeepSeek | deepseek-flash |
Not exposed by the current Nova adapter | DEEPSEEK_API_KEY |
| Qwen / DashScope | qwen-plus |
Not on the default model; select a listed Qwen vision model | DASHSCOPE_API_KEY |
| Volcengine Ark | doubao-seed-2-0-pro-260215 |
Supported | ARK_API_KEY |
| OpenAI | gpt-6-luna |
Not exposed by the current Nova adapter | OPENAI_API_KEY |
| Gemini | gemini-3.5-flash-lite |
Not exposed by the current Nova adapter | GEMINI_API_KEY |
| Self-hosted (OpenAI-compatible endpoint) | None; set CASCADE_LLM_MODEL |
Not exposed by the current Nova adapter | Optional SELF_HOSTED_LLM_API_KEY |
A model-name override does not grant image capability. A model supporting images is separate from Nova having a verified image-input adapter for it. Tool use also requires compatibility with the adapter's structured tool protocol.
Vision models
| Provider | Models recognized by Nova |
|---|---|
| Qwen / DashScope | qwen3-vl-plus, qwen3-vl-flash, qwen-vl-max, qwen-vl-plus |
| Volcengine Ark | doubao-seed-2-0-pro-260215 |
Conversation vision uses the selected cascaded language model. Loop Camera uses a separate monitor-model connection, so an audio-only or DeepSeek foreground conversation can still delegate monitoring to a vision model. Unknown models do not silently fall back to another model.
Speech recognition and synthesis
| Stage | Current adapter | Default resource / voice |
|---|---|---|
| ASR | Volcengine Speech (default); Gemini (CASCADE_ASR_PROVIDER=gemini); self-hosted streaming ASR (self-hosted) |
volc.seedasr.sauc.duration; Gemini gemini-3.5-flash |
| TTS | Volcengine Speech (default); Gemini (CASCADE_TTS_PROVIDER=gemini); self-hosted TTS (self-hosted) |
seed-tts-2.0 / zh_female_vv_uranus_bigtts; Gemini gemini-3.8-flash-tts / Kore |
| Integrated voice | Qwen realtime | longanqian |
Self-hosted stages talk to endpoints you run (SELF_HOSTED_ASR_URL, SELF_HOSTED_LLM_BASE_URL, SELF_HOSTED_TTS_URL) and use only their own optional credentials; the repository ships a reference launcher for a streaming Whisper ASR, a vLLM LLM and a Breeze TTS. Cloud and self-hosted stages can be mixed in one pipeline. See self-hosted voice and presets.
For Volcengine, ASR uses DOUBAO_ASR_API_KEY when present, otherwise DOUBAO_BIGMODEL_API_KEY. TTS uses the latter. Supporting models and personal memory may still require DashScope credentials; see configuration.
Voiceprint verification
| Setting | Behavior |
|---|---|
| Opt-in | Off by default; enable with DOUBAO_ASR_VOICEPRINT_ENABLED |
| Registered speaker | DOUBAO_ASR_VOICEPRINT_ID and DOUBAO_ASR_VOICEPRINT_NAME identify the enrolled voiceprint |
| Health fallback | Verification disables itself (fails open) when DOUBAO_ASR_VOICEPRINT_HEALTH_URL is unreachable |
Implementation: voiceprint.
Connectors
| Connector | Access | Platform |
|---|---|---|
| Google (Gmail, Calendar) | Read, via Composio | Cross-platform |
| macOS Calendar | Read | macOS only |
| Apple Mail | Read | macOS only |
| Feishu | Read-only | Cross-platform |
See sources and connectors for setup and scopes.
Implementation: defaults, cascaded adapter selection, vision allowlist.