iTANONG
A DOST-ASTI system for querying databases in Filipino, English or Taglish
Overview
iTANONG is a natural-language interface to databases built by the Philippines' DOST-ASTI since 2022, letting users ask questions in Filipino, English or Taglish instead of writing SQL. It began as a three-year research project, is now supported by the national ACABAI-PH AI programme, and is described by DOST-ASTI as built for secure on-premises use in government. It is not open to the public.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 19 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
What you can produce with iTANONG
- Questions to a relational database typed in Filipino, English or Taglish, answered without SQL
- On-premises deployment for government agencies handling internal data
- Open Tagalog and Cebuano BERT, RoBERTa, BigBird and GPT-2 models from DOST-ASTI on Hugging Face
- A Tagalog sentiment-analysis model and Taglish fastText word embeddings
- The iTANONG-DS paper describing Philippine-language NLP benchmarks
- Partnership routes for universities and agencies through DOST-ASTI agreements
ASEAN Perspective
iTANONG in Southeast Asia
While major ASEAN economies have state- or conglomerate-backed language models, the Philippines has had less activity. iTANONG, from DOST-ASTI under the ACABAI-PH programme, is the most visible public-sector effort to change that. It lets users query databases in English, Tagalog and Taglish without SQL — Taglish being the point, since code-switching is how Filipinos actually write and is precisely what off-the-shelf models mangle. Set expectations accordingly: this is an early-stage government research project begun in 2022, not a product you can adopt.
iTANONG tackles a real problem: letting Filipinos who are not programmers get answers out of government databases in the mix of Filipino and English they actually speak. Its by-products are useful in their own right, notably open Tagalog and Cebuano language models and Taglish embeddings on Hugging Face.
But there is nothing to adopt. Access is restricted, no user numbers or independent evaluation exist, and DOST-ASTI describes it as an on-premises system for government. Agencies should contact DOST-ASTI directly; everyone else should treat it as a research record.
What we found
iTANONG is a Philippine government research system, not a product you can sign up for. It was started in January 2022 by the Department of Science and Technology's Advanced Science and Technology Institute (DOST-ASTI) to let people question relational databases in Filipino, English or Taglish instead of SQL. Its project page sets the run at January 2022 to December 2024 and lists government offices, banks, retailers, schools and disaster agencies as hoped-for adopters.
The engineering is mixed rather than homegrown from the ground up. A July 2023 Rappler feature reported that the team built on OpenAI's GPT models combined with in-house models, and used synthetic data and scraping of Filipino sources because Filipino is a low-resource language; normalising spellings and street slang was part of the work. DOST-ASTI published Tagalog and Cebuano BERT, RoBERTa, BigBird and GPT-2 models on Hugging Face between November 2023 and March 2024, plus Taglish word embeddings. The team's 2023 iTANONG-DS paper describes benchmark datasets as part of a 10-billion-token initiative, but only the embeddings, not those labelled datasets, appear under DOST-ASTI's Hugging Face account.
Adoption has been by agreement rather than by download. DOST-ASTI signed a memorandum with National University and Bicol University in May 2023 to promote it. In October 2023 the institute's director told Philstar Tech that a mobile version, more languages such as Cebuano and Ilokano, and a speech interface were being explored, with no launch date set.
It has outlived its original end date. In May 2025 DOST-ASTI listed iTANONG among the platforms supported by ACABAI-PH, its national AI programme launched that February; earlier versions of this page reversed that order. An April 2026 programme report calls it a generative AI interface for querying government databases, designed for secure on-premises deployment. Its web address answers with a 403. No user numbers or independent evaluation have been published.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
Frequently asked questions
About this listing
This entry was compiled from publicly available data including iTANONG's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with iTANONG unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to iTANONG directly →
Spotted something out of date? Suggest an update →
Alternatives to iTANONG
More in LLMs & Chat
