# Deepgram updates Every Deepgram update in one feed — release notes, features and changes, straight from the source. 52 updates - ## [Deepgram adds two voice models and welcomes Daily]() Think-Live runs the conversation and is generally available. Think-Fast does the work and is in early access. Daily, the company behind Pipecat, is joining Deepgram. 2026-10-29 - ## [Verify Callback Requests with Basic Auth or Extra Metadata]() Deepgram callbacks no longer support dg-token verification; use Basic Auth for STT, TTS and Text Intelligence, or an Extra Metadata secret for STT. 2026-10-27 - ## [Nova-3 Multilingual Expands to 55 Languages and Adds Language Hinting]() Requests using model=nova-3&language=multi get the new model automatically, and an optional language\_hint parameter steers detection toward expected languages. 2026-10-23 - ## [Japanese Numerals on Multilingual Models]() Numerals now formats Japanese on Nova-3 and Nova-3 Medical multilingual and Flux Multilingual, while Hindi stays unsupported. 2026-10-23 - ## [Flux TTS Adds 41 Languages and 230 Voices]() Flux TTS adds voices in languages from Bengali to Vietnamese on the same endpoints, with text normalization in nine and Early Access for the other 32. 2026-10-23 - ## [Deepgram Self-Hosted October 2026 Release (261020)]() Deepgram's 261020 self-hosted images add FIPS 140-3 variants and a new Flux runtime that no longer loads old .dg model files. 2026-10-20 - ## [Confirm Nova-3 Configure Updates]() Every Configure message on a Nova-3 /v1/listen stream now gets a ConfigureSuccess or ConfigureFailure reply, with the keyterms in effect. 2026-10-19 - ## [Update Nova-3 Keyterms Mid-Stream on Every Regional Endpoint]() Nova-3 Configure messages on /v1/listen now work on the EU, Australia and India endpoints, and behave the same as on the global endpoint. 2026-10-19 - ## [Python SDK v7.13.0]() Python developers can update live Nova-3 streams without reconnecting and exchange experimental custom Think-provider messages. 2026-10-12 - ## [JavaScript SDK v5.15.0]() Developers can update Nova-3 keyterms mid-stream and exchange experimental custom Think-provider messages in Voice Agent sessions. 2026-10-12 - ## [Go SDK v3.9.0]() Go developers get typed Voice Agent function calls and management endpoints for reusable configurations and template variables. 2026-10-12 - ## [.NET SDK v7.1.3]() Listen V1 WebSocket connections now accept an omitted model, with or without keyterms. 2026-10-12 - ## [Rust SDK v0.13.0]() The release adds text analysis and caption helpers, with breaking changes to usage fields and custom base URL handling. 2026-10-12 - ## [Rust SDK v0.12.0]() Streaming fixes prevent hangs and expose terminal transport errors, while sends on ended /v1/listen sessions now return Err. 2026-10-12 - ## [Deepgram CLI v0.3.3]() New flags play generated speech and list voices, while skill installs use a pinned, hash-checked v1.7.0 bundle. 2026-10-12 - ## [Java SDK v0.11.1]() Java developers can update Nova-3 keyterms mid-stream, handle typed server errors and exchange custom Think-provider payloads. 2026-10-12 - ## [Deepgram skills v1.7.0]() All 14 skills refresh API, SDK, and CLI guidance for coding assistants including Claude Code, Cursor, and Codex. 2026-10-12 - ## [Earlier releases: These releases, from September 14 through October 5]() The SDK updates add mid-stream Numerals controls, Flux TTS pause markers and Early Access pronunciation overrides. 2026-10-12 - ## [Correction: over-limit Configure keyterm updates on /v1/listen]() Exceeding 500 tokens stops Nova-3 transcription without an Error, then closes the stream after about 30 seconds. 2026-10-08 - ## [Nova-3 Medical: Improved English Models for Batch and Streaming]() Hosted users receive the improvements automatically; self-hosted customers must request updated models from their account representative. 2026-10-07 - ## [Nova-3 Medical Multilingual: Dutch, English, French, German, Hindi, Italian, Japanese, Portuguese, Russian, and Spanish]() Hosted and self-hosted customers can use the API for both pre-recorded medical audio and real-time streaming. 2026-10-06 - ## [Nova-3 Improved Models for Catalan, French, Indonesian, and Vietnamese]() The updates apply automatically to existing API requests; self-hosted customers must request updated models from their account representative. 2026-10-05 - ## [Update Nova-3 Keyterms Mid-Stream]() Nova-3 streaming users can replace keyterms with a Configure message on the global endpoint; EU, Australia and India endpoints aren't supported yet. 2026-10-02 - ## [Deepgram Self-Hosted October 2026 Release (261001)]() Flux TTS audio is now watermarked, and self-hosted Engine will not start with Flux TTS enabled until the watermarker file and UUID are configured. 2026-10-01 - ## [Flux TTS Inline Controls: Pause and Pronunciation]() Flux TTS adds inline pauses on batch requests and Early Access IPA pronunciation overrides on batch and streaming, with errors for invalid combinations. 2026-09-30 - ## [Toggle Numerals Mid-Stream on Flux STT]() Voice app developers can toggle digits versus words on a live Flux STT stream, on the English model and eight languages of the multilingual one. 2026-09-25 - ## [Nova-3 Improved Models for Flemish, German (Switzerland), Lithuanian, and Portuguese]() Nova-3 transcription improves for German (Switzerland), Portuguese, Lithuanian and Flemish, with no API changes required. 2026-09-24 - ## [Nova-3 Improved Models for Danish, Estonian, Flemish, Italian, Lithuanian, Macedonian, Polish, Urdu, and Vietnamese]() Nine Nova-3 language models get quality updates for batch or streaming transcription, live now with no API changes required. 2026-09-22 - ## [Correction: Browser Agent SDK has no client-side VAD]() Deepgram says client-side Silero VAD was documented but never shipped in the Browser Agent SDK; barge-in runs server-side instead. 2026-09-18 - ## [Correction: token minting uses ttl\_seconds]() Deepgram fixes a docs example that silently ignored the token lifetime developers tried to set for Voice Agent API calls. 2026-09-18 [Open Deepgram updates on Creators Toolbox]() --- Canonical source: https://creatorstoolbox.com/updates/deepgram