Cohere Transcribe
Cohere Transcribe is a 2B parameter, open-source, automatic speech recognition model developed by Cohere and Cohere Labs, supporting 14 languages. It is genuinely best at transcribing speech to text, utilizing a conformer-based encoder-decoder architecture.
Input
Output
Context
33K
Max Output
-
Parameters
2.1B
Technical Specifications
Model TypeAudio
Context Window32,768 tokens
Max Output TokensNot available
Parameters2.1B
Release DateMar 24, 2026
Training CutoffNot available
Licenseapache-2.0
Open SourceYes
Input Modalities
Audio
Output Modalities
Text
Capabilities
Resources & Links
Estimated GPU Requirements for Cohere Transcribe
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Jun 23, 2026
Start building with the right model.
Automatically route workloads to the right model for every task, every time.