Skip to content

chore(main): release 6.0.0 - #765

Closed
github-actions[bot] wants to merge 1 commit into
mainfrom
release-please--branches--main--components--deepgram-sdk
Closed

chore(main): release 6.0.0#765
github-actions[bot] wants to merge 1 commit into
mainfrom
release-please--branches--main--components--deepgram-sdk

Conversation

@github-actions

Copy link
Copy Markdown
Contributor

🤖 I have created a release beep boop

6.0.0 (2026-08-12)

⚠ BREAKING CHANGES

  • sdk regeneration 2026-04-24 (#696)
  • sdk regeneration 2026-04-14 (#690)
  • promote v6.0.0-rc.2 to v6.0.0 stable
  • v6 — fully generated SDK with latest APIs and WebSocket support (#640)
  • This is a significant breaking change, and should be carried out in conjunction with our migration guide.

Features

  • (regen) diarization v2 batch GA, listen-provider dedup, think/models route fix (#714) (9d9a43d)
  • add declarative reconnect flag with transport-factory auto-disable (#720) (b5d5905)
  • Add dictation option to LiveOptions class (8939e59)
  • Add dictation option to LiveOptions class (d3b2dcc)
  • add projects billing fields list methods (#621) (10d67cd)
  • add support for agent tags (#559) (d605248)
  • adds support for agent mip_opt_out (#558) (3e8a3dd)
  • Adds support for IUM (#546) (232ee05)
  • adds support for speak fallback (#548) (cab41fa)
  • adds ttl_seconds support (#557) (e9bedf2)
  • agent: support multi-provider speak/think configuration and typed listen parameters (#676) (5dfb1aa)
  • apply latest spec changes (4ad9eb7)
  • auth: adding token-based auth to the SDK so access tokens can be used in requests (#543) (1200c83)
  • helpers: add TextBuilder class for TTS pronunciation and pause controls (#660) (4324120)
  • implements new generated SDK architecture, all call signatures (#572) (768d514)
  • make all providers generic (2c92d67)
  • make provider objects generic (4521048)
  • mention keep alive in migration guide (#594) (5a8c79e)
  • officially enable aura-2 in live text-to-speech (052aadd)
  • officially enable aura-2 in live text-to-speech (ba4124a)
  • promote v6.0.0-rc.2 to v6.0.0 stable (34f543e)
  • regen: add Flux TTS batch (REST) endpoint and agent latency report (#744) (6ca71c4)
  • regen: add Speak V2 streaming, agent update-listen, and Flux EOT tuning (#742) (69a2445)
  • regen: flux stt numerals and aura-2 multilingual tts voices (#746) (17a1deb)
  • regen: listen v2 force-end-turn and redaction, listen v1 diarize metadata (#750) (4678f0e)
  • regen: speak v2 barge-in and speed/expressivity controls, listen v2 redaction (#758) (aab1eae)
  • regen: v2 language_hints, profanity filter, word timings, diarize_model (#730) (da6b7ba)
  • sagemaker: add optional extras group for SageMaker dependencies (#659) (2046175)
  • SDK regeneration (21 Oct 2025) (#609) (5b21460)
  • sdk regeneration 2026-04-14 (#690) (d4d129f)
  • sdk regeneration 2026-04-24 (#696) (4714207)
  • support arbitrary attributes for speak provider (1973bbc)
  • support context length option (14ea4bf)
  • support context length option (3735cc0)
  • support for agent v1 (5945c56)
  • support for agent v1 (7ad48ea)
  • support short-lived tokens endpoint (016bd4d)
  • support short-lived tokens endpoint (3474a88)
  • update generated SDK models and restore agent settings compatibility (#705) (0b820c9)
  • update the SDK with upstream spec changes (d77ad96)
  • updateinstructions to updateprompt (a269dc7)
  • v6 — fully generated SDK with latest APIs and WebSocket support (#640) (bc918fe)
  • websockets: add custom WebSocket transport support (#658) (f6cf0fb)

Bug Fixes

  • 🌿 skip_validation:true to allow unknown messages back from the API (#669) (48354d2)
  • (regen) route think/models to env.agent_rest (#715) (ffd2e7d)
  • adds inject agent message method (#555) (c0e864b)
  • agent: move mip_opt_out field to root level of SettingsOptions (#561) (eca802a)
  • ci: reference the correct secret (#585) (09550c7)
  • coderabbit gave some good tips this time (eb03638)
  • correct the language property location and type (6380b56)
  • corrects order to the release workflow (#583) (3abbac3)
  • couple more issues (8e066ce)
  • default value (72e61a2)
  • do coderabbit's stuff (d062caf)
  • downgrade 'tasks cancelled error' to debug log (#550) (eca4b1e)
  • endpoint class fix (#551) (1cc054f)
  • function calls (#556) (20fb81c)
  • handle empty objects for providers (a2d66fe)
  • handle empty objects for providers (eddcf18)
  • handle socket connection closed error in _signal_exit (5dee108)
  • handle socket connection closed error in _signal_exit (2ca6dbd)
  • issue of exception on websocket closing (bdbc10a)
  • lint (d647d63)
  • lowercase bool query params on websocket connect (#712) (8899609)
  • moves agent tags to settings (4fc8756)
  • moves agent tags to settings (e15781d)
  • redact Authorization header from websockets DEBUG logs (#731) (029d877)
  • release-please config fixes (#579) (a603806)
  • remove testpypi we don't need it in the workflow (#582) (b2e2538)
  • remove the keyterms check (4ab6cba)
  • resolve lint issue (b21005a)
  • sagemaker: extract SageMaker transport to separate deepgram-sagemaker package (#665) (e6317c5)
  • shut up linter (4cf0542)
  • speak: correct TTS warning event field names to match API response (#653) (f7ab1da)
  • support multiple keyterms for v2 listen client (#595) (7a9d41d)
  • types: change speaker and related fields from float to int (#652) (00ee485), closes #641
  • update error handling in async_client.py and client.py (4f78e8e)
  • updates examples to aura-2 [WIP] (#514) (e48e415)
  • websockets: restore optional message param on control send_ methods (#680) (0018fc4)
  • websockets: support array parameters in Listen v1 and v2 clients (#650) (38cc1e5)
  • widen pydantic-core bound via fern 5.14.8 regen (closes #701) (#724) (5c1e845)

Reverts

Documentation

  • improve contributor onboarding and setup instructions (#718) (3f7398a)
  • target Context7 benchmark gaps in Python skills [no-ci] (#699) (a232eb8)

Miscellaneous Chores


This PR was generated with Release Please. See documentation.

@GregHolmes

Copy link
Copy Markdown
Contributor

Superseded: v7.7.0 released and published. This PR computed a bogus version (6.0.0) from a missing-tag state; release-please will open a correct next-release PR on the next push to main.

@GregHolmes GregHolmes closed this Aug 12, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[Bug]: pydantic-core = ">=2.18.2,<2.44.0" ListenV1ResponseResultsUtterancesItem.speaker defined as float

1 participant