Private-beta on-device speech generation delivered as a licensed binary tailored to a voice and deployment. Supports five languages and fully offline inference.
Private beta by application. Provider builds and licenses a partner-specific binary for the target devices and voice.
Regions
Inference runs on the customer device.
Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms
Offline inference keeps runtime text on the device. Voice samples and fine-tuning/build inputs supplied to Gradium have separate contractual handling; offline inference does not establish zero retention for training uploads.
Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.
01 / INITIAL INPUTProcess & retain
The system processes eligible input and retains reusable state.
02 / LATER REQUESTCheck available state
Reuse depends on matching input, session state and the cache policy.
03 / CACHE HITReuse & generate
Reusing state can reduce repeated work. Any charges follow the billing rules below.
Licensed on-device deployment
Not verified
No reviewed model-specific prompt-cache contract. Reused WebSockets, stored voices and pronunciation dictionaries do not establish discounted prompt caching.
How it works
Not collected
No reviewed model-specific prompt-cache contract. Reused WebSockets, stored voices and pronunciation dictionaries do not establish discounted prompt caching.
Minimum reusable input
Not collected
No reviewed model-specific prompt-cache contract. Reused WebSockets, stored voices and pronunciation dictionaries do not establish discounted prompt caching.
Lifetime & refresh
Not collected
No reviewed model-specific prompt-cache contract. Reused WebSockets, stored voices and pronunciation dictionaries do not establish discounted prompt caching.
How to enable it
Not collected
No reviewed model-specific prompt-cache contract. Reused WebSockets, stored voices and pronunciation dictionaries do not establish discounted prompt caching.
Usage fields
Not collected
No reviewed model-specific prompt-cache contract. Reused WebSockets, stored voices and pronunciation dictionaries do not establish discounted prompt caching.
Billing rules
Not collected
No reviewed model-specific prompt-cache contract. Reused WebSockets, stored voices and pronunciation dictionaries do not establish discounted prompt caching.
Limits & data boundaries
Not collected
No reviewed model-specific prompt-cache contract. Reused WebSockets, stored voices and pronunciation dictionaries do not establish discounted prompt caching.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
What the measurements say
No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.
Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.
The record behind the model
Why it is included
The proprietary service license reserves ownership of models and restricts copying, redistribution and extraction. Phonon is supplied as a separately licensed, partner-specific binary; offline deployment is not an open-weight license. Public SDKs and evaluation datasets do not grant production model weights.
