The Open Weights
LatestModelsLeaderboardsCompanies
Subscribe
The Open Weights

The daily record of open-source AI. New model releases, leaderboards, and what's coming next — written for people who ship.

Refreshed every 12 hours

Discover

  • Latest releases
  • New today
  • Trending models

Browse

  • All models
  • Companies
  • Categories
  • Leaderboards

About

  • About
  • Editorial policy
  • Hardware estimates
  • RSS feed
  • llms.txt
  • Newsletter

© 2026 The Open Weights. An independent publication.

PrivacyTermsSMSAggregated by Claude · curated by humans.
LatestMicrosoftStreaming-7B
MicrosoftSpeech → Text

Microsoft's VibeVoice ASR brings streaming speech-to-text

A 7-billion-parameter model targets real-time, multilingual transcription with an open release on Hugging Face.

Sep 2, 2026
NotableOther
VibeVoice-ASR-Streaming-7B

Microsoft has published VibeVoice-ASR-Streaming-7B, a speech recognition model designed for streaming transcription, on Hugging Face. The 7-billion-parameter dense model handles automatic speech recognition across multiple languages, including English, Chinese, and Spanish.

The "streaming" designation is the key detail here. Rather than waiting for a full audio clip to finish before producing text, streaming ASR emits transcriptions incrementally as audio arrives. That makes the model a better fit for live captioning, meeting notes, voice interfaces, and any application where latency matters as much as accuracy.

Why it matters

Speech-to-text has become a crowded field, but most of the attention has gone to batch-oriented models. A capable open-weight streaming option from a major lab gives developers more flexibility to build low-latency, on-device or self-hosted transcription pipelines.

  • 7B parameters, dense (not a mixture-of-experts model)
  • Multilingual coverage spanning English, Chinese, Spanish, and more
  • Built specifically for streaming, real-time transcription

The model is distributed under a custom license, so teams evaluating it for production should review the terms on the model card before deploying. As an initial release in the VibeVoice ASR line, it also signals Microsoft's continued investment in open speech tooling.

Sources

  • microsoft/VibeVoice-ASR-Streaming-7B

    Hugging Face

    Visit
OlderRWKV7-G1j arrives as a 13.3B attention-free modelFla Hub · Text / LLM · 2 months agoNewerLLaDA-Image: A Fully Open 6B Image GeneratorinclusionAI · Text → Image · 2 months ago

Get the model

Hugging Face

Specs

Parameters7B
Languagesen, zh, es +
Size17.3 GB
PrecisionBF16
ArchitectureVibeVoiceForASRStreamingTraining
LicenseOTHER
Downloads10.9K
Likes268

Can you run it?

Runs on a laptop — about 9.7 GB at FP8.

  • BF16 (as published)

    24 GB GPU (RTX 3090 / 4090) · 32 GB Mac

    18.3 GB
  • FP8

    12 GB GPU (RTX 3060 / 4070) · 16 GB Mac

    9.7 GB
Your machine

Based on Hugging Face weights metadata. Assumes an 8K context and ~1 GB runtime overhead; actual needs vary. How we estimate


Modalities

Speech → Text

The Weekly Weights

Every open release that mattered, one email a week.

0 comments

No comments yet. Be the first to weigh in.

More from Microsoft

All Microsoft releases →
Mage-VL-Cowork
Microsoft/Vision-LanguageRuns on a laptop

Microsoft's Mage-VL-Cowork takes on computer-use tasks

A new vision-language agent model aims to operate GUIs, write code, and work the terminal.

Oct 30, 2026
FrogNano-4B
Microsoft/Text / LLMRuns on a laptop

Microsoft Releases FrogNano-4B Under MIT License

The compact 4-billion-parameter text model lands on Hugging Face with a permissive license aimed at developers.

Sep 17, 2026
Mage-VL
Microsoft/Vision-LanguageRuns on a laptop

Microsoft's Mage-VL Streams Video Natively

A codec-native multimodal foundation model aims to understand live video and vision-language input in real time.

Jul 26, 2026

More in Speech → Text

All Speech → Text →
Haru-4B
Higurashi Ai/Any-to-AnyRuns on a laptop

Haru-4B Brings Full-Duplex Speech to Six Languages

A compact 4-billion-parameter model handles real-time, two-way voice conversation across six languages in a single system.

Nov 12, 2026
ServiceNow/Text → SpeechConsumer GPU

ServiceNow Debuts Apriel Voice 27B, an Open Speech Model

The 27-billion-parameter preview unifies text-to-speech, transcription, and reasoning in a single open model.

Oct 28, 2026
IBM/Speech → Text

IBM's Granite Speech 5.0 Adds Speaker Attribution

The Apache-2.0 speech model handles transcription, translation, and figuring out who said what.

Oct 21, 2026