GPTViet: An Open-Source Vietnamese Foundation Model from Pretraining to Domain Specialization
Bibliographic record
Abstract
As open-source Large Language Models (LLMs) increasingly rival proprietary counterparts, the need for foundational models tailored to specific linguistic and cultural contexts becomes critical. This paper presents GPTViet, a series of foundational LLMs for the Vietnamese language. Built upon the LLaMA architecture, GPTViet was developed by curating a high-quality Vietnamese corpus and performing extensive finetuning on a range of model sizes (8 B to 70 B parameters). Evaluations demonstrate that GPTViet models significantly outperform their respective base models on Vietnamese-specific tasks, as measured by standard benchmarks and our custombuilt VietExam benchmark. The practical utility of this work is showcased through domain-specific application, VietHealth 70B for medical consultation. Adhering to the principles of open source Llama, GPTViet and all its derivatives are publicly released under an open-source license. This initiative provides the Vietnamese research and development community with a powerful, adaptable foundation to accelerate the creation of diverse intelligent applications. For more information, please visit https://github.com/VietnamAIHub/GPTViet and demo at http://gptviet.ioit.ac.vn/.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.014 | 0.006 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".