md-nishat-008/TigerLLM-1B-it

Hugging Face
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:1BQuant:BF16Context Size:32kPublished:Sep 9, 2025License:cc-by-4.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Warm

TigerLLM-1B-it is a 1 billion parameter instruction-tuned causal language model from the TigerLLM family, developed by Nishat Raihan and Marcos Zampieri at George Mason University. Built on the Gemma 3 architecture, this compact model is specifically optimized for instruction-following in Bangla, capable of answering questions, explaining, rewriting, and summarizing. Its primary use case is providing efficient and accurate Bangla language processing, even on resource-constrained devices like laptops or CPUs, with a context length of 32,768 tokens.

Loading preview...

TigerLLM-1B-it: A Compact Bangla Instruction-Tuned LLM

TigerLLM-1B-it is a 1 billion parameter instruction-following model built on the Gemma 3 architecture, developed by Nishat Raihan and Marcos Zampieri. This model is part of the TigerLLM family, presented at ACL 2025, and is specifically designed for efficient Bangla language processing. It is notable for its compact size, allowing it to run on laptop GPUs or even CPUs, while maintaining strong performance in Bangla instruction-following tasks.

Key Capabilities

  • Bangla Instruction Following: Excels at understanding and responding to instructions in Bangla.
  • Multifaceted Bangla Tasks: Capable of answering questions, providing explanations, rewriting text, and summarizing content in Bangla.
  • Resource-Efficient: Designed to be light enough for deployment on devices with limited computational resources, such as laptops or CPUs.
  • Extended Context: Features a context length of 32,768 tokens, enabling processing of longer inputs.
  • High-Quality Training Data: Fine-tuned on the Bangla-Instruct dataset, comprising 100,000 native Bangla instruction-response pairs, ensuring high relevance and quality for Bangla tasks.

Good For

  • Bangla Language Applications: Ideal for developers building applications that require robust instruction-following capabilities in Bangla.
  • Edge Device Deployment: Suitable for scenarios where models need to run efficiently on consumer-grade hardware.
  • Research and Development: A valuable resource for researchers exploring compact, high-performance LLMs for low-resource languages.
  • Educational Tools: Can be integrated into tools for learning or content generation in Bangla.

Finetunes
1 models