Store › model › Speech
granite-speech-4.1-2b-nar-gguf
by handy-computer · source Hugging Face · updated 2026-09-15
apache-2.01.5 GB~3 GB RAMsource aliveunlabeled
granite-speech-4.1-2b-nar: transcribe.cpp GGUF
Add to LogiShell Open in the web IDE
The button opens LogiShell with this card; nothing installs from a link by itself. Inside the app the install goes through lsh models install hf:handy-computer/granite-speech-4.1-2b-nar-gguf and its progress lives in the Resource Center.
Source and license
- Source: https://huggingface.co/handy-computer/granite-speech-4.1-2b-nar-gguf
- License: apache-2.0
- Requirements: about 3 GB of RAM, 1.5 GB on disk (estimate: suggested file size × 1.15 + 0.5 GB; a real measurement comes with lsh models). Runs with whisper.cpp.
- Tags:
transcribe.cppggufasrspeech-to-textgranitegranite-speechnarnon-autoregressivemultilingualautomatic-speech-recognitionenfrdeespt
Numbers
- 39,084 downloads on Hugging Face
- 0 likes
- license apache-2.0
- 1.5 GB for granite-speech-4.1-2b-nar-Q4_K_M.gguf
- 15,195 npm downloads a week for nodejs-whisper
- latest nodejs-whisper@0.3.1
Numbers as of 2026-10-02 21:00 UTC, from the source API.
Summary
Summary not ready yet: the numbers are here, the text is not. It is written by the collector through the LogiShell model facade when a provider key is present.
Reviews
No reviews yet. Reviews are written inside LogiShell: open this card in the app.
Files
| file | quant | size |
|---|---|---|
granite-speech-4.1-2b-nar-BF16.gguf | BF16 | 4.2 GB |
granite-speech-4.1-2b-nar-F16.gguf | F16 | 4.2 GB |
granite-speech-4.1-2b-nar-Q4_K_M.gguf | Q4_K_M | 1.5 GB |
granite-speech-4.1-2b-nar-Q5_K_M.gguf | Q5_K_M | 1.7 GB |
granite-speech-4.1-2b-nar-Q6_K.gguf | Q6_K | 1.8 GB |
granite-speech-4.1-2b-nar-Q8_0.gguf | Q8_0 | 2.3 GB |
From the source README
granite-speech-4.1-2b-nar: transcribe.cpp GGUF
GGUF conversions of ibm-granite/granite-speech-4.1-2b-nar for use
with transcribe.cpp.
Ported from upstream commit
99a4df9,
pinned 2026-05-24.
Validated against the Transformers reference at transcribe.cpp commit
c53af2c
on 2026-05-24.
Offline multilingual speech-to-text in a single non-autoregressive editor
pass. IBM Granite Speech 4.1-2b NAR shares the Conformer audio encoder
with the AR Granite-Speech family but pairs it with a custom MLP-with-
attention projector and the Granite-4.0-1b LLM used as a bidirectional
editor (causal mask disabled). One forward pass produces logits over the
full transcript; CTC decode yields the final text. No token-by-token loop.
Takes a 16 kHz mono WAV and produces a transcript. English plus French,
German, Spanish, and Portuguese; ASR only (no translation, no
timestamps).
Downloads
| Quantization | Download | Size | WER (LibriSpeech test-clean) |
| --- | --- | ---: | ---: |
| BF16 | granite-speech-4.1-2b-nar-BF16.gguf | 4.51 GB | 1.29% |
| F16 | granite-speech-4.1-2b-nar-F16.gguf | 4.52 GB | 1.29% |
| Q8_0 | granite-speech-4.1-2b-nar-Q8_0.gguf | 2.50 GB | 1.29% |
| Q6_K | [granite-speech-4.1-2b-nar-Q6_K.gguf](https://huggingface.co/handy-computer/granite-speech-4.1-2b-nar-gguf/resolve/mai…
Source: https://huggingface.co/handy-computer/granite-speech-4.1-2b-nar-gguf
Card id model:hf:handy-computer/granite-speech-4.1-2b-nar-gguf · collected 2026-10-02 21:00 UTC · JSON