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.
LatestReson82-Flash
Reson8Speech → Text

Reson8's Rumble-2-Flash Puts ASR in 0.5B Params

A compact open speech recognition model aims to transcribe more than 40 languages without heavyweight infrastructure.

Nov 18, 2026
NotableOther
Rumble-2-Flash

Reson8 has released Rumble-2-Flash, an open speech-to-text model that packs multilingual automatic speech recognition into roughly 500 million parameters. According to the company's announcement on Hugging Face, the model handles transcription across more than 40 languages.

The headline here is size. At 0.5B parameters, Rumble-2-Flash sits well below the billion-parameter threshold, which makes it a candidate for deployment on modest hardware, edge devices, or latency-sensitive pipelines where larger ASR systems are impractical.

Why it matters

Multilingual speech recognition has historically demanded significant compute, and the strongest systems have often been closed or expensive to run. A small, openly available model that spans dozens of languages lowers the barrier for developers building:

  • Offline or on-device transcription
  • Voice interfaces for underserved languages
  • Cost-constrained batch processing at scale

The model is published under a non-standard license, so teams should review the terms before building on it commercially. As the first entry in the Rumble-2 line, it establishes a baseline worth watching as Reson8 iterates on coverage and accuracy.

Sources

  • Rumble-2-Flash: an open 0.5B speech model for 40+ languages

    Announcement

    Visit
  • Reson8/Rumble-2-Flash

    Hugging Face

    Visit
OlderSenseTime Releases Looped 9B Multimodal ModelSenseTime · Vision-Language · 3 days agoNewerMicrosoft's Aion 1.0 Instruct is a compact 2.3B chat modelMicrosoft · Text / LLM · 2 days ago

Get the model

Hugging FaceAnnouncement

Specs

Parameters500M
Languages40+ languages
Size1 GB
PrecisionBF16
ArchitectureQwen3ASRForConditionalGeneration
LicenseOTHER
Downloads1.8K
Likes181

Can you run it?

Runs on just about anything.

  • BF16 (as published)

    Any modern laptop · 8 GB RAM

    2 GB
  • FP8

    Any modern laptop · 8 GB RAM

    1.5 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 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