NTQAI/NxMobileLM-1.5B-SFT

Hugging Face
TEXT GENERATIONPricing:Input $0.04 / Cached $0.008 / Output $0.08Concurrent Unit Cost:1Model Size:1.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jan 15, 2025License:mitArchitecture:Transformer0.0K Open Weights Featherless Exclusive Warm

NTQAI/NxMobileLM-1.5B-SFT is a 1.5 billion parameter language model, fine-tuned from Qwen2.5-1.5B, specifically optimized for efficient performance in mobile and edge computing environments. It excels in natural language understanding and generation tasks, demonstrating strong multilingual capabilities across various benchmarks. This model is primarily designed for lightweight, real-time language-based applications such as mobile virtual assistants and compact edge AI solutions.

Loading preview...

NxMobileLM-1.5B-SFT Overview

NTQAI's NxMobileLM-1.5B-SFT is a 1.5 billion parameter language model built upon the Qwen2.5-1.5B base, specifically engineered for mobile and edge device deployment. It has undergone supervised fine-tuning (SFT) using proprietary instruction datasets to enhance its performance in natural language understanding and generation, with a strong emphasis on efficiency and multilingual support.

Key Capabilities

  • Optimized for Edge Devices: Designed for low-latency inference and efficient operation on mobile and edge hardware.
  • Multilingual Proficiency: Demonstrates strong performance across multiple languages, including English, Japanese, and Vietnamese, making it suitable for global applications.
  • Instruction Following: Fine-tuned on diverse instruction datasets for tasks like question answering, summarization, and dialogue.
  • Competitive Performance: Achieved top rankings on the Open-LLM Leaderboard for models under 2 billion parameters and shows strong results on benchmarks like P-MMEval and LightEval, particularly in instruction-following (IFEval) and mathematical reasoning (GSM8K).

Good For

  • Mobile Virtual Assistants: Powering intelligent assistants on smartphones and other portable devices.
  • Real-time Language Applications: Scenarios requiring quick natural language processing directly on the device.
  • Compact Edge AI Solutions: Deploying AI capabilities in resource-constrained environments.
  • Multilingual Scenarios: Applications needing robust cross-lingual communication and understanding.

Finetunes
1 models