Who should use GigaAM?

GigaAM is an open-source project built for speech technology developers and researchers.

Yes — GigaAM is open source under the MIT license and free to use.

What platforms does GigaAM run on?

GigaAM runs on the web, API, and the command line. It can also be self-hosted.

Is GigaAM still maintained?

The GitHub repository shows 4 commits in the last 90 days.

GigaAM is developed by ai-sage.

Is GigaAM open source?

Yes — GigaAM is open source under the MIT license.

Back to the index

GigaAM

Name: GigaAM
Author: ai-sage

huggingface.coInfrastructure

GigaAM-v3 is an open-source Conformer-based foundation model for automatic speech recognition, focused primarily on Russian speech. It offers self-supervised, CTC, RNN-T, and end-to-end variants with punctuation and text normalization for developers and researchers.

Open SourceMITWebSelf-hostedAPICLI

Visit huggingface.co

765stars

102forks

1alternative

9features

2024since

Overview

9 features

In the Voice, TTS & speech space, GigaAM takes a focused approach. It focuses on transcribing Russian and English speech into text with punctuation and normalization. It is built as an open-source project for speech technology developers and researchers. The project is open source (MIT). It runs on the web, API, and the command line, and it can be self-hosted.

Behind GigaAM is ai-sage, and it first shipped in 2024. Development happens publicly on GitHub with 765 stars and 4 commits in the last 90 days. Key capabilities include automatic speech recognition, russian speech support, and english speech support.

Summary written by a language model from the project’s public pages.

Automatic speech recognition
Russian speech support
English speech support
Self-supervised encoder
CTC decoding
RNN-T decoding
Punctuation restoration
Text normalization
PyTorch models

Built with & integrations

Hosting

AWS

Runs on

BrowserSelf-hostedAPI-onlyCLI

Detected from

AWS: x-amz-cf-id header · x-amz-cf-pop header · via header

Trust & compliance

License

MIT

Verified signals

HTTPSPrivacy Policy Terms of ServiceOpen SourceActive maintenance

Legal

Indexing history

What PulseGate has recorded for this listing

Indexed26 Aug · 15:23 UTC
ai-sage/GigaAM-v3 verified via Hugging Face Enumerator
Source: Hugging Face Enumerator · Open

Frequently asked questions about GigaAM

What is GigaAM?: GigaAM focuses on transcribing Russian and English speech into text with punctuation and normalization. It is catalogued under Voice, TTS & speech on PulseGate.
Who should use GigaAM?: GigaAM is an open-source project built for speech technology developers and researchers.
Is GigaAM free?: Yes — GigaAM is open source under the MIT license and free to use.
What platforms does GigaAM run on?: GigaAM runs on the web, API, and the command line. It can also be self-hosted.
Is GigaAM still maintained?: The GitHub repository shows 4 commits in the last 90 days.
Who develops GigaAM?: GigaAM is developed by ai-sage.
How long has GigaAM been around?: GigaAM first shipped in 2024.
Is GigaAM open source?: Yes — GigaAM is open source under the MIT license.

At a glance

Pricing: Open Source
Platforms: Web
Languages: English
License: MIT
Built for: speech technology developers and researchers
Model: Open source
Solves: Transcribing Russian and English speech into text with punctuation and normalization.

Registered as

Hugging Face: ai-sage/GigaAM-v3

Developer

ai-sage

Team

Open source

Stars: 765
Forks: 102
Open issues: 15
Last commit: 17 Aug 2026
Commits 90d: 4
Contributors: 6
Authorship: Team
Default branch: main

Index record

Indexed: 26 Aug 2026
Lifecycle: Alive
Last seen: 26 Aug 2026

Identity audit (13)

Slug: ai-sage-gigaam-v3-huggingface-co
Lifecycle last checked: 8 Sept 2026
Verification state: Indexed for public listing
Listing state: Listed: yes
Index status: Included in index
Latest evidence snapshot: 26 Aug 2026
Timeline basis: Indexed-at chronology. This listing's first-seen date was written by the catalog backfill, not observed here, so it is not treated as a sighting.
Name from: Derived from the project's own page and URL.
Category from: Derived from the model's own Hugging Face metadata.
Summary from: Written by a language model from public pages.
Languages from: Detected by a language model from page content.
Last updated: 26 Aug 2026
Canonical URL: https://huggingface.co/ai-sage/GigaAM-v3

Ship this? Send a correction — no account, and you get a link to follow it.

Similar projects

Closest matches by what these projects do