Releasesby GlyphNova-3 Improved Models for Flemish, German (Switzerland), Lithuanian, and Portuguese
We have released improved Nova-3 monolingual models for four existing languages. These updates enhance transcription quality for batch and streaming workloads.
Also shipped Sep 24, 2026Deepgram in September 2026 →
DCM Projects capability updates
This release updates the availability of several DCM Projects capabilities. The following capabilities are now generally available: The following capabilities are now available in preview:
Alpha Changelog - 9/23/26
Hi everyone, here are the latest round of changes on alpha.midjourney.com . Many of these came directly from your feedback in #alpha-ideas-and-bugs . Please keep it coming! STYLE PREVIEWS We're experimenting with fast models in the interface.
New project activity feed
As part of our ongoing work to simplify the Netlify dashboard, we’ve added an activity feed to the project overview page. It’s available on all plans and adapts to how your project is set up.
Retool CLI
Live on cloud • Team • Business • Enterprise Public beta 4.62 Edge • Team • Business • Enterprise Public beta 4.34.4 Stable • Team • Business • Enterprise Public beta The Retool CLI is now available in public beta.
Buildkite Cache is now in public preview
Buildkite Cache is now available in public preview. It saves and restores files and directories across pipeline jobs and builds. Use it for package manager download caches, compiled dependencies, and other regenerable data to avoid repeating work.
Service principal Genie usage is billed at standard pricing
Genie usage by service principals is now billed at standard Genie pay-as-you-go pricing. Service principals are excluded from the free monthly usage allowance that applies to individual users. See Manage budgets and cost controls for Genie and Monitor and understand your Genie cost .
More from DeepgramEvery Deepgram release →
Improved models
Batch and streaming models German (Switzerland) ( de-CH ) Portuguese ( pt , pt-BR , pt-PT ) Lithuanian ( lt ) Streaming models Flemish ( nl-BE ) You do not need to change your API requests to access the improvements to existing languages; the updates are live for all users.
Toggle Numerals Mid-Stream on Flux STT
Switch Flux STT to digits for a PIN, phone number, or order number, then back to words, without reconnecting. To turn Numerals on or off during a Flux STT stream, send numerals as a boolean in a Configure message: { "type": "Configure", "numerals": true } The update applies to transcripts Flux STT sends after it processes the message, and ConfigureSuccess...
Nova-3 Improved Models for Danish, Estonian, Flemish, Italian, Lithuanian, Macedonian, Polish, Urdu, and Vietnamese
We have released improved Nova-3 monolingual models for nine existing languages. These updates enhance transcription quality for batch and streaming workloads.
Improved models
Batch and streaming models Polish ( pl ) Danish ( da ) Batch models Vietnamese ( vi ) Flemish ( nl-BE ) Streaming models Estonian ( et ) Macedonian ( mk ) Urdu ( ur ) Italian ( it ) Lithuanian ( lt ) You do not need to change your API requests to access the improvements to existing languages; the updates are live for all users.
Flux TTS Inline Controls: Pause and Pronunciation
Flux TTS ( /v2/speak ) now supports inline pause and pronunciation controls in the text you send, alongside speed . Pause control. Insert silence with an escaped marker such as \{pause:1s\} : 500–3000 ms in 100 ms steps, up to 8 per request. Available on batch (REST) requests. Pronunciation control (Early Access).
Correction: Browser Agent SDK has no client-side VAD
The Browser Agent SDK never shipped client-side Silero voice activity detection. Documentation published with the SDK described it as an available feature, and that documentation was wrong. The following did not exist in any released package: The vad option on MicrophoneOptions in @deepgram/agents . The speechThreshold and silenceThreshold VAD options.
Correction: token minting uses ttl_seconds
The server-side token-minting example on the Browser Agent SDK overview posted {"ttl": 30} to POST /v1/auth/grant . The field is named ttl_seconds : body: JSON.stringify({ ttl_seconds: 30 }), /v1/auth/grant accepts unrecognized fields, returns 200 , and issues a token with the 30-second default, so a request built from the old example raised no error while...
Container Images (release 261001)
- quay.io/deepgram/self-hosted-api:release-261001 Equivalent image to: quay.io/deepgram/self-hosted-api:1.192.51 quay.io/deepgram/self-hosted-engine:release-261001 Equivalent image to: quay.io/deepgram/self-hosted-engine:3.135.0 Minimum required NVIDIA driver version: >=580 , open kernel module flavor.