Releasesby GlyphDeprecation Notice
The Gemini 2.5 Flash family of models is deprecating in October. Start testing newer models now to ensure a smooth migration. For more details on Gemini model deprecations, see Google's Gemini deprecations page . For the full list of supported models and pricing tiers, see the Voice Agent LLM Models documentation .
Also shipped May 27, 2026Deepgram in May 2026 →
Fedora 42 has reached end of life
Fedora 42 has reached end of life. Per our image deprecation policy, this image is available exclusively via the API for the next 30 days before we remove it from our platform.
AI-Detected Issues Now Generally Available
Sentry can use AI to analyze your traces and logs to detect issues that traditional pattern-matching would miss, like database inefficiencies, blocking operations, and security concerns.
Next Generation Events now available in developer preview
Next Generation Events are now available in developer preview. Events give you field-level control over when a subscription fires, what data it carries, and what triggers each delivery, fixing three common friction points faced with traditional webhooks: over-delivery, fixed payloads, and no built-in signal for what changed.
Frustration Analytics , Debugger on User Profile Timeline
What: The User Profile Timeline now surfaces frustration signals directly alongside behavioral events – error events with distinct styling, key properties inline, error-focused filtering, and quick links to Session Replay. This gives you a chronological view of exactly what a specific user experienced.
GitHub CLI 2.93.0
Security A security vulnerability has been identified, and fixed, that would incorrectly include authorization header in API requests to TUF repository mirrors via gh attestation , gh release verify , and gh release verify-asset commands. Users are advised to update gh to version v2.93.0 as soon as possible.
Web Updates
Conversational mode has been improved across text and voice input: When a voice session starts, it has access to your Image Prompts, Style References, sidebar settings, and recent jobs. Image Prompts now work from the tray and sidebar. Tray images persist across voice submissions until you remove them. You can add references without exiting voice.
More from DeepgramEvery Deepgram release →
Gemini 3.5 Flash Now Available
gemini-3.5-flash is now available as a managed Google LLM in the Voice Agent API. This Standard tier model brings improved performance and efficiency to your voice agents. Set the model in your agent configuration: { "agent": { "think": { "provider": { "type": "google", "model": "gemini-3.5-flash", "temperature": 0.5 } } } }
Container Images (release 260528)
- quay.io/deepgram/self-hosted-api:release-260528 Equivalent image to: quay.io/deepgram/self-hosted-api:1.188.1 quay.io/deepgram/self-hosted-engine:release-260528 Equivalent image to: quay.io/deepgram/self-hosted-engine:3.117.0 Minimum required NVIDIA driver version: >=570.172.08 quay.io/deepgram/self-hosted-license-proxy:release-260528 Equivalent image to...
Preparation for a future Engine container change
The official Helm chart ( 0.37.0 and later) and the Docker and Podman compose files in deepgram/self-hosted-resources now set NVIDIA_VISIBLE_DEVICES=all and NVIDIA_DRIVER_CAPABILITIES=compute,utility on the Engine container.
This Release Contains The Following Changes
- Profanity Filtering for STT Multilingual , profanity_filter=true now masks recognized profanity in STT multilingual transcripts ( language=multi ). See Profanity Filtering for the supported language list and usage.
🆕 Improved Nova-3 Medical Batch Model Released
We’ve released an upgraded Nova-3 Medical batch model with improved medical term recognition. Key Improvements: Expands the model’s medical vocabulary with new medical terms Boosts medical term recognition to 97.20% KRR Maintains WER parity with no observed regression on common or conversational terms
🆕 Profanity Filtering for Multilingual Models
Deepgram’s Profanity Filtering feature is now available for all multilingual models: Nova-2 multilingual , Nova-3 multilingual , and Flux multilingual ( language=multi ). You can enable profanity filtering in your API requests by setting the profanity_filter=true parameter.
🛠️ Fix: Improved Word Spacing in Korean Transcripts
We fixed an issue affecting Korean transcripts ( ko , ko-KR ) where word spacing was sometimes missing. Transcripts should now better reflect proper Korean spacing, improving readability for users working with Korean audio. See the full list of supported languages on the Models & Languages Overview page .
Gemini 3.1 Flash Lite Now Available
gemini-3.1-flash-lite is now available as a managed Google LLM in the Voice Agent API. This Standard tier model replaces the preview version. Set the model in your agent configuration: { "agent": { "think": { "provider": { "type": "google", "model": "gemini-3.1-flash-lite", "temperature": 0.5 } } } }