Reson8 has open-sourced its 0.5B speech recognition model Rumble-2-Flash
The story so far
Reson8 has open-sourced its 0.5B parameter speech recognition model Rumble-2-Flash under the Apache 2.0 license, supporting over 40 languages and automatically detecting language. The model weights are available on the Hugging Face Hub under Reson8/Rumble-2-Flash and can be accessed via API calls. According to the Hugging Face blog, the model natively supports Transformers and vLLM without custom code; it can control the output to be Arabic numerals or text, padding tokens, character repetition, and force language specification; with full batch processing, it exceeds 7,000 RTFx, processing one hour of audio in about half a second. The blog claims its performance can compete with models 3 to 60 times its parameter size.
AI-written from the reports · Updated 6 hours ago
Coverage timeline
Follow the coverage to see every side of the story.
- Hugging Face BlogFeaturedRumble-2-Flash: an open 0.5B speech model for 40+ languages
Reson8 has open-sourced Rumble-2-Flash under the Apache 2.0 license, which is a 0.5B parameter speech recognition model supporting over 40 languages and automatically detecting the language, natively supporting Transformers and vLLM.
Heat over time
- Now
- 6
- Peak
- 718 Nov, 23:00
- 24h change
- –
Hover or tap the chart to see each hour; use the arrow keys on a keyboard.