Project AURIONAI Use-Cases Research & Innovation
IBM
speech recognition

granite-4.0-1b-speech

IBM Granite 4.0 speech recognition model.

IBMIBMUSA
Word error rate chart of granite-4.0-1b-speech from the model card.
Levelmodel
Classspeech recognition
Parameters2.31B
Contextnot in the notes
Released27 Feb 2026
Opennessopen weights permissive
Runs onsingle datacentre GPU
Statusactive

Benchmark Results

maintainer-run
Hugging Face Open ASR Leaderboardspeech
eight-set macro-average English WERmaintainer-run
5.52
048

Score in WER. Lower is better.

snapshot 2026-05-06 via benchmarklist6 May 2026 benchmarklist.com
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).

Lineage

successor
granite-4.0-1b-speech to granite-speech-4.1-2b: successor, medium confidencegranite-speech-3.3-8b to granite-speech-4.1-2b: successor, medium confidencegranite-speech-3.3-8bgranite-speech-3.3-8bIBM, 2025granite-4.0-1b-speechgranite-4.0-1b-speechIBM, Feb 2026granite-speech-4.1-2bgranite-speech-4.1-2bIBM, Apr 2026
IBMgranite-speech-4.1-2b is a child (successor)medium confidencenotes
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 3 models. Confidence and evidence are the notes' own.

R&D Topics

Massively multilingual speech recognitionproduced by

Identity

Developer
IBM
Family
not in the notes
Released
27 Feb 2026 sourcesregistry date
Weights published
–
Record checked
10 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
autoregressive
Total parameters
2.31B sources
Active parameters
–
Checkpoint parameters
2.31B
Note
1B by name; HF reports 2.31B total, conflicts with name (unverified)
Context length
not in the notes
Input
audio
Output
text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights permissive
Licence
Apache License 2.0 sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
single datacentre GPU
Footprint
not in the notes
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Industry Use Cases

No industry use case links this model.

Sources

  1. Open ASR Leaderboard Benchmark Scores & AI Model Leaderboard | BenchmarkList benchmarklist.com https://benchmarklist.com/benchmarks/open_asr_leaderboard/
  2. huggingface.co/api/models/ibm-granite/granite-4.0-1b-speech https://huggingface.co/api/models/ibm-granite/granite-4.0-1b-speech

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Granite-4.0-1b-Speech", https://aurion.semai.id/models/ibm-granite-4-0-1b-speech/, records of October 2026.