Skip to content

granite-speech-4.1-2b-nar-mlx-5bit

mlx-community/granite-speech-4.1-2b-nar-mlx-5bit
5-bit Open on Hugging Face
Parameters
2.3B
Weights (est.)
≈ 2.4 GB
Mac memory (est.)
8 GB+

Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.

Details

Base model
ibm-granite/granite-speech-4.1-2b-nar (quantized)
Vendor
IBM · Granite
Task
Speech recognition
Quantisation
5-bit
Licence
apache-2.0
Downloads
51 in the last 30 days · 402 all time · 0 likes
Published
2026-05-28 · updated 2026-05-28 (5mo ago)
Languages
en, fr, de, es, pt
Tags
granite_speech_nar, mlx-audio, speech-to-text, non-autoregressive, granite, custom_code

Other MLX builds of ibm-granite/granite-speech-4.1-2b-nar

Build Quantisation Weights Mac memory Downloads
granite-speech-4.1-2b-nar-mlx-5bit (this one) 5-bit ≈ 2.4 GB 8 GB+ 51
granite-speech-4.1-2b-nar-mlx — ≈ 4.5 GB 8 GB+ 217