Score: 2

TigerCoder: A Novel Suite of LLMs for Code Generation in Bangla

Published: September 11, 2025 | arXiv ID: 2509.09101v1

By: Nishat Raihan, Antonios Anastasopoulos, Marcos Zampieri

Potential Business Impact:

Helps computers write computer code in Bangla.

Business Areas:
Natural Language Processing Artificial Intelligence, Data and Analytics, Software

Despite being the 5th most spoken language, Bangla remains underrepresented in Large Language Models (LLMs), particularly for code generation. This primarily stems from the scarcity of high-quality data to pre-train and/or finetune such models. Hence, we introduce the first dedicated family of Code LLMs for Bangla (1B & 9B). We offer three major contributions: (1) a comprehensive Bangla code instruction datasets for programming domain adaptation; (2) MBPP-Bangla, an evaluation benchmark for Bangla code generation; and (3) the TigerCoder-family of Code LLMs, achieving significant ~11-18% performance gains at Pass@1 over existing multilingual and general-purpose Bangla LLMs. Our findings show that curated, high-quality datasets can overcome limitations of smaller models for low-resource languages. We open-source all resources to advance further Bangla LLM research.

Country of Origin
🇺🇸 United States

Repos / Data Links

Page Count
17 pages

Category
Computer Science:
Computation and Language