Post-Cutoff
Search For AI agents

DeepgramFlux familyCurrentSpeech and audioReleased 2 Oct 2025

Deepgram Flux (conversational STT, English + Multilingual)

STT built for voice agents: transcription + end-of-turn detection in one model.

Model idflux-general-en

Price$0.0065 per minute

Checked29 September 2026

Where it stands

82newer speech and audio models since its release

The 5 newest are listed. All speech and audio models

Before it from Deepgram
Deepgram Aura-215 Apr 2025
After it from Deepgram
Deepgram Flux TTS12 Aug 2026
Weights
Closed weights; access is hosted only.

How to call it

2 ways to use Deepgram Flux (conversational STT, English + Multilingual), with the model id each provider expects.

Providers, model ids, endpoints and docs for Deepgram Flux (conversational STT, English + Multilingual)
ProviderModel idEndpointDocs
Deepgram API flux-general-en wss://api.deepgram.com/v2/listen developers.deepgram.com
Deepgram API flux-general-multi wss://api.deepgram.com/v2/listen developers.deepgram.com

Pricing

Prices for Deepgram Flux (conversational STT, English + Multilingual)
Per minute$0.0065

As published: “per minute streaming, pay-as-you-go (flux-general-en); Flux Multilingual $0.0078/min; Growth plan $0.0057 / $0.0068”.

Source: deepgram.com, checked 29 September 2026.

Specs

Input
Audio
Output
Text
Context window
Not published
Max output
Not published
Open weights
No
Licence
Proprietary
Released
2 October 2025
Status
Current
Type
Speech and audio

What stands out

What its maker marketed at launch, and what people found later, each with its source.

  • Conversational speech recognition with model-native turn-taking

    First

    Recognition model itself decides end-of-turn using acoustic + semantic cues (~260 ms end-of-turn detection), with EagerEndOfTurn events to start the LLM early; tunable eot_threshold, eager_eot_threshold, eot_timeout_ms.

    Source: deepgram.com

  • Multilingual conversational STT with in-call code-switching

    FirstFound after launch

    Flux Multilingual (GA 2026-04-29): English, Spanish, French, German, Hindi, Russian, Portuguese, Japanese, Italian, Dutch with automatic language switching mid-conversation; turn detection under 400 ms. Billed by Deepgram as the world’s first multilingual conversational speech recognition model.

    Source: deepgram.com

Notes

‘First’ claims are Deepgram’s own marketing (launched at VapiCon 2025-10-02 as ‘world’s first conversational speech recognition model’). Uses /v2/listen (not /v1). Mid-stream numeral toggle added 2026-09-25. Companion TTS: deepgram-flux-tts.

Sources: https://developers.deepgram.com/docs/flux/quickstart , https://deepgram.com/pricing , https://developers.deepgram.com/changelog