ibm-ai-platform/micro-g3.3-8b-instruct-1b

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Jun 26, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

ibm-ai-platform/micro-g3.3-8b-instruct-1b is a 1-billion parameter micro language model developed by ibm-ai-platform. Built on the Granite-3.3-8B-Instruct architecture with only 3 hidden layers, it is fine-tuned for reasoning and instruction-following. This model is optimized to maximize performance and hardware compatibility at minimal compute cost, making it suitable for efficient deployment.

Loading preview...

Model Overview

Micro-G3.3-8B-Instruct-1B is a compact yet powerful 1-billion parameter instruction-tuned language model from ibm-ai-platform. It is based on the Granite-3.3-8B-Instruct architecture but features a significantly reduced hidden layer count (only 3), making it a 'micro' model.

Key Capabilities

  • Reasoning: Designed to handle complex reasoning tasks effectively.
  • Instruction-Following: Excels at understanding and executing user instructions.
  • Efficiency: Optimized for high performance and broad hardware compatibility with minimal computational overhead.

Use Cases

This model is particularly well-suited for applications where:

  • Resource Constraints: Compute resources are limited, requiring a highly efficient model.
  • Edge Deployment: Deployment on devices with restricted memory and processing power.
  • Cost-Sensitive Operations: Minimizing inference costs is a priority.
  • Instruction-Based Tasks: Applications requiring robust instruction adherence and logical reasoning.