amjad0099/gemma3-cidar-ar

TEXT GENERATIONConcurrent Unit Cost:1Model Size:0.3BQuant:BF16Context Size:32kPublished:Jul 9, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The amjad0099/gemma3-cidar-ar is a 0.3 billion parameter Gemma3-based causal language model, fine-tuned by amjad0099. This model was trained using Unsloth and Huggingface's TRL library, enabling faster fine-tuning. With a context length of 32768 tokens, it is suitable for tasks requiring efficient processing of longer sequences.

Loading preview...

Model Overview

The amjad0099/gemma3-cidar-ar is a 0.3 billion parameter language model, fine-tuned by amjad0099. It is based on the Gemma3 architecture, specifically fine-tuned from the unsloth/gemma-3-270m-it model. This model leverages the Unsloth library and Huggingface's TRL library for accelerated training, indicating an optimization for efficient fine-tuning processes.

Key Characteristics

  • Architecture: Gemma3-based, a causal language model.
  • Parameter Count: 0.3 billion parameters, making it a compact model.
  • Context Length: Supports a substantial context window of 32768 tokens.
  • Training Efficiency: Fine-tuned using Unsloth, which facilitates 2x faster training.

Potential Use Cases

Given its compact size and efficient fine-tuning, this model is well-suited for:

  • Applications requiring a smaller footprint and faster inference.
  • Tasks benefiting from a long context window, such as summarization or detailed question answering.
  • Further experimentation and fine-tuning on specific downstream tasks where rapid iteration is beneficial.