Project AURIONAI Use-Cases Research & Innovation
LLM

llama3-8b-cpt-sahabatai-v1

Sahabat-AI v1 continued-pretraining model (base and instruct).

GoTo Gojek TokopediaGoTo Gojek TokopediaIndonesiaIndosat Ooredoo HutchisonIndosat Ooredoo HutchisonIndonesiaAI SingaporeAI SingaporeSingaporeSahabat-AI v1Family
GoTo Gojek Tokopedia
Levelmodel
ClassLLM
Parameters8B
Contextnot in the notes
Released6 Nov 2024
Opennessopen weights restricted
Runs onconsumer GPU
Statusactive

Lineage

continued pretrain
Llama 3 8B to llama3-8b-cpt-sahabatai-v1: continued pretrain, high confidenceLlama 3 8B to LLama-3-8B-Tele-it: continued pretrain, high confidenceLlama 3 8BLlama 3 8BMetaLLama-3-8B-Tele-itLLama-3-8B-Tele-itSep 2024llama3-8b-cpt-sahabatai-v1llama3-8b-cpt-sahabatai-v1GoTo Gojek Tokopedia, Nov 2024
MetaLlama 3 8B is the parent (continued pretrain)high 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.

Industry Use Cases

Public Sector
Citizen service assistantsIndonesian and regional-language LLMs

R&D Topics

Low-resource and regional language resourcesproduced byRegional or national foundation model training (continued pretraining at 7B to 70B)produced by

Deployments

Indosat Ooredoo Hutchison
Indosat Ooredoo HutchisonproductionIndonesia13 Nov 2024

Answers citizen questions such as ID card updates and tax rates; platform for the government sovereign AI initiative

Citizen service assistantscompany statementblogs.nvidia.com
Where this comes from
From G13, the deployment rows of T08 to T10, and the Indonesian-language search of 11 October 2026. An outcome claim is the organisation's own unless the evidence type says independent report.

Identity

Developer
GoTo Gojek Tokopedia, Indosat Ooredoo Hutchison, AI Singapore
Family
Sahabat-AI v1
Released
6 Nov 2024 sourcesnotes
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
not in the notes
Total parameters
8B sources
Active parameters
–
Checkpoint parameters
–
Context length
not in the notes
Input
text
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 restricted
Licence
Meta Llama 3 Community License sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
consumer GPU
Footprint
Open weights; on-prem or Indonesian territory; sovereign cloud GPU Merdeka by Lintasarta
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
listed by developer
Regional languages
Javanese, Sundanese, Batak Toba, Balinese
Jakarta hosting
unknown
Languages
Indonesian + regional languages
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Benchmark Results

The notes record no benchmark result for this model.

Sources

  1. GoToCompany (PT GoTo Gojek Tokopedia Tbk) huggingface.co https://huggingface.co/GoToCompany
  2. Models – Hugging Face huggingface.co https://huggingface.co/models?search=sahabat
  3. Indonesia Tech Leaders Team With NVIDIA and Partners to Launch Nation’s AI blogs.nvidia.com https://blogs.nvidia.com/?p=75648

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "llama3-8b-Cpt-Sahabatai-v1", https://aurion.semai.id/models/goto-llama3-8b-cpt-sahabatai-v1/, records of October 2026.