G
GRADIUM / gradium/phonon-31m-en

Phonon 31M English

Gradium text-to-speech model with 31 million parameters, excluding the audio tokenizer. Published on October 19, it added its Apache-2.0 license file on November 17. It generates English speech with 64.5 MB q8 GGUF weights.

01 / SUMMARY

The essential

ORIENTATIONOn-device English speech synthesis

Primary use stated or inferred cautiously from official documentation.

CONTEXTEnglish text; maximum length unpublished

Maximum input capacity when published by the source.

EXITMono audio at 24 kHz, generated in streaming

Documented maximum generation limit.

ACCESSHugging Face · downloadable checkpoint

Verified availability channels.

MODEL TYPEAudio

Primary functional family and verified specialties.

LICENSEApache-2.0

Declared terms for API access, model weights, or self-hosting.

02 / TECHNICAL SHEET

Limits and integration

API IDgradium/phonon-31m-en
Model typeAudio
Access modelDownloadable weights
LicenseApache-2.0
DeploymentSelf-hosted
Release19 OCT 2026
Knowledge cutoffNot published
EntranceText
ExitAudio
Context windowEnglish text; maximum length unpublished
maximum outputMono audio at 24 kHz, generated in streaming
ReasoningNot published
Published tools
Structured OutingsNot published
Batch processingNot published
Prompt cacheNot published
Fine-tuningNot published
Verified platformsHugging Face
BEST SUITED FOR

Local speech synthesis in a browser, phone or CPU. It uses a flow-matching transformer and the Mimi decoder to produce 80 ms audio frames. The license covers weights, tokenizer and voices in this repository.

WORTH MONITORING

This profile covers only the 31M English model. The 40M variants use other licenses and do not inherit Apache-2.0. Audio frames of 80 ms do not guarantee that latency on every device. Review the runtime and permissions for additional voices.

03 / CAPABILITIES

What it can do

01

Orientation

Local speech synthesis in a browser, phone or CPU. It uses a flow-matching transformer and the Mimi decoder to produce 80 ms audio frames. The license covers weights, tokenizer and voices in this repository.

02

Context

English text; maximum length unpublished · Mono audio at 24 kHz, generated in streaming

03

Tools and integration

Not published

04

Access

Hugging Face · downloadable checkpoint

Input modalities
Text
04 / PRICES

Documented cost

ConceptWorthUnit/condition
Standard inputNo published official API rateStandard API rate
Cached inputNot publishedReading reused prefixes
Cache write or storageNot publishedThe condition varies by provider
Standard outputSelf-hosted infrastructure costMay include reasoning tokens
Batch inputNot publishedAsynchronous processing
Batch outputNot publishedAsynchronous processing

The repository publishes Apache-2.0, including commercial use under its terms. The runtime is separately licensed MIT or Apache-2.0. No official API price is verified; local execution consumes your own resources.

i Prices change and may depend on level, region or context length. Check the source before making a decision.

05 / EVALUATIONS

How to read the results

“

The card reports English US Elo 2098 from 468 raters and 8,290 votes. This is a manufacturer listening-preference comparison, not an independent measurement or a pronunciation guarantee for every text.

BenchmarkResultMetricSource
Phonon English US listening Elo2098Elo · listening preferenceView source ↗

Inferama only highlights a “best result” when the metric, test set, configuration, and date allow for an equivalent comparison. The supplier's figures are presented as claims from its own source.

06 / VERSIONS

Chronology

gradium/phonon-31m-en

Original weights publication.

8f7a37c9

Apache-2.0 file added to the repository; the model’s original date is unchanged.

07 / SOURCES

Traceability