Universal-3.6
Q4 2026A lower-cost Universal-3.6 streaming model at $0.30/hr, alongside Universal-3.6 Pro.
What's coming next across the AssemblyAI stack — from transcription and understanding through speech, conversation, and the agent platform — and what shipped in the last 90 days.
Speech-to-text across pre-recorded, realtime, and sync audio — the foundation everything else is built on.
A lower-cost Universal-3.6 streaming model at $0.30/hr, alongside Universal-3.6 Pro.
An option to keep filler words, false starts, and repetitions exactly as spoken, instead of a cleaned-up transcript.
Now available as universal-3-6-pro-preview; GA follows evaluation.
Available now in preview via domain: veterinary-v1 (English, Spanish,
French, German). GA in Q4.
The next major accuracy and latency release after Universal-3.6 Pro, across pre-recorded and realtime audio, with the strongest handling of voice-agent audio (noise, interruptions, hesitation, accented speech).
Finance and legal modes, following the Medical Mode pattern already in production — domain vocabulary and formatting tuned per vertical.
domain: medical-v1 that cuts missed medical
terms by up to 13% against the base model, on pre-recorded and streaming audio in
English, Spanish, German and French.
speech_model is sent. 32
languages with native mid-sentence code-switching (14 of them new), entity-aware
endpointing, and better handling of background voices. Same $0.45/hr as
Universal-3.5 Pro. New languages: Galician, Korean, Catalan, Russian, Afrikaans,
Romanian, Cantonese, Marathi, Persian, Estonian, Urdu, Norwegian Nynorsk, Zulu,
Xhosa.
language_codes on Universal-3.5 Pro
Pre-recorded language_codes accepts any set of supported languages
— English no longer has to be included.
speaker_options.include_speaker_confidence to get a
confidence score for the speaker assigned to each word and utterance.
format_text: false on Universal-3.5 Pro pre-recorded audio to get
transcripts in spoken form, without written-form formatting.
Turning transcripts into meaning: LLM inference over audio, speaker intelligence, and the guardrails that keep output safe to ship.
Ongoing catalog expansion across frontier and open-weight models, added as they release.
Send audio directly to LLM Gateway models that accept it, alongside text.
Speaker identification improvements, including support for multichannel audio.
Recognize the same speaker across recordings, not just within a single file. For meetings, call centers, and cross-session analytics.
Category redaction for entity types you define, building on the static find-and-replace redaction that has already shipped.
Ask plain-language yes/no questions about a recording — is the caller frustrated, is this a voicemail, did the agent read the disclosure — and get back calibrated probabilities instead of text. The model listens to tone and delivery as well as the words, so decisions that a transcript alone can't make become a single API call.
Our own LLM, fine-tuned for voice agents and transcription use cases.
/v1/models endpoint lists every model with
its pricing.
Text-to-speech: the voices your application speaks with.
Wider language coverage matching the speech-to-text model lineup, audio watermarking, and prosody improvements including no leading breath before a sentence.
Full-duplex conversation — voice agents, dictation, and live translation.
Research toward a native speech-to-speech model for voice agents, aiming for tighter latency, better prosody, and more natural interruption handling — without giving up control over turn-taking.
How models run in production — across transports, regions, and deployment options.
Rate limits that scale automatically by provider, region, and model instead of needing a manual bump, starting in October. Visible token-per-minute limits and automatic failover away from a provider that is down follow.
Self-hosted streaming and sync are available today; November adds Universal-3.6 Pro and standalone speaker diarization. Pre-recorded (async) self-hosting is not planned.
LLM Gateway service tiers (priority, standard, and flex) for trading latency against cost.
Our models running on the device, for offline and latency-critical workloads where audio never leaves the endpoint.
modal deploy.
Building, deploying, and operating voice agents on AssemblyAI.
Official client libraries that own the connection and audio pipeline, so agents work on the first try. The Python SDK launches in October; TypeScript follows.
One-click deploy of Voice Agent SDK agents to AssemblyAI’s cloud, so you don’t run your own agent infrastructure. Live today with an early customer.
The dashboard, accounts, and tooling that make AssemblyAI easy to adopt — for humans and AI agents.
Hard spend caps and team budgets, beyond today’s soft alerts.
Programmatic access to billing and usage data.
Deeper dashboard observability: P50 and P95 turnaround time, webhook delivery stats, uptime, and latency histograms.
mcp.assemblyai.com/mcp to submit, read, summarize and analyze
transcripts. Authorize with Sign in with AssemblyAI or an API key.