Changelog

Deepgram Self-Hosted November 2026 Release (261110)

Container Images (release 261110)

• quay.io/deepgram/self-hosted-api:release-261110

  • Equivalent image to:
    • quay.io/deepgram/self-hosted-api:1.192.58-1

• quay.io/deepgram/self-hosted-engine:release-261110

  • Equivalent image to:

    • quay.io/deepgram/self-hosted-engine:3.142.0-1

  • Minimum required NVIDIA driver version: >=580, open kernel module flavor. See Drivers and Containerization Platforms for installation and verification steps.

• quay.io/deepgram/self-hosted-license-proxy:release-261110

  • Equivalent image to:
    • quay.io/deepgram/self-hosted-license-proxy:1.11.2-1

• quay.io/deepgram/self-hosted-billing:release-261110

  • Equivalent image to:
    • quay.io/deepgram/self-hosted-billing:1.15.1-1

FIPS 140-3 Images

FIPS 140-3 images are available for all four self-hosted services, published under the -fips tag suffix:

• quay.io/deepgram/self-hosted-api:release-261110-fips

  • Equivalent image to:
    • quay.io/deepgram/self-hosted-api:1.192.58-1-fips

• quay.io/deepgram/self-hosted-engine:release-261110-fips

  • Equivalent image to:
    • quay.io/deepgram/self-hosted-engine:3.142.0-1-fips

• quay.io/deepgram/self-hosted-license-proxy:release-261110-fips

  • Equivalent image to:
    • quay.io/deepgram/self-hosted-license-proxy:1.11.2-1-fips

• quay.io/deepgram/self-hosted-billing:release-261110-fips

  • Equivalent image to:
    • quay.io/deepgram/self-hosted-billing:1.15.1-1-fips

These images run OpenSSL’s FIPS 140-3 validated cryptographic provider. Enable FIPS by setting [fips] mode = "enabled" in each service’s configuration. FIPS mode should be enabled only on the -fips images; the standard images do not include the FIPS provider. See FIPS-Compliant Deployment for setup and configuration details.

This Release Contains The Following Changes

• Nova-3 Multilingual Language Hints (action required) — a language_hint on a multilingual model that does not support hints now returns HTTP 400 INVALID_QUERY_PARAMETER. Previously the hint was accepted but had no effect. Before upgrading, remove language_hint from requests to multilingual models that don’t support hints.
• Flux STT Language Hint Validation — an unsupported language_hint on /v2/listen now returns HTTP 400 when the stream is opened, and a Configure message with one returns ConfigureFailure and keeps the stream’s existing configuration. Previously the stream failed after the first audio.
• Callback Verification — the dg-token header is no longer a supported method to verify that callback requests come from Deepgram. The API still sends it in this release, but if your callback server checks for dg-token, switch to Basic Auth by putting credentials in the callback URL, such as https://username:password@example.com/callback. For Speech-to-Text, you can also pass a secret with Extra Metadata, such as extra=callback_secret:YOUR_SECRET_VALUE, and reject any callback whose metadata.extra.callback_secret is missing or doesn’t match. See Authenticating Callback Requests for details.
• Preloaded Models Use Half Precision — on GPUs that support it, models loaded at startup through [preload_models] now load in half precision, matching models loaded on demand. This reduces GPU memory use and improves inference speed for preloaded models.
• Flux TTS Graceful Shutdown — Engine now lets in-progress /v2/speak WebSocket sessions finish before shutting down, matching /v2/listen.
• Formatting Improvements — Portuguese “meia” is now rendered as 6 inside spelled-out digit strings.
• General Improvements — License Proxy and Billing rebuilt with updated dependencies. Keeps our software up-to-date.