Hothaifa/HEQ-S2-1.6.6

VISIONConcurrent Unit Cost:2Model Size:31BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 18, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Hothaifa/HEQ-S2-1.6.6 is a 31 billion parameter Gemma 4 model, fine-tuned by Hothaifa, offering a 32768-token context length. This model was trained using Unsloth and Huggingface's TRL library, enabling 2x faster training. It is designed for general language tasks, leveraging its efficient fine-tuning process for improved performance.

Loading preview...

Model Overview

Hothaifa/HEQ-S2-1.6.6 is a 31 billion parameter language model, fine-tuned by Hothaifa. It is based on the Gemma 4 architecture and supports a substantial context length of 32768 tokens. This model was developed using an efficient fine-tuning process, leveraging the Unsloth library and Huggingface's TRL library, which reportedly enabled training at twice the speed.

Key Characteristics

  • Architecture: Gemma 4 base model.
  • Parameter Count: 31 billion parameters.
  • Context Length: Supports up to 32768 tokens.
  • Training Efficiency: Fine-tuned with Unsloth, resulting in 2x faster training.
  • License: Released under the Apache-2.0 license.

Good For

  • Applications requiring a large context window for processing extensive text.
  • General language generation and understanding tasks where the Gemma 4 architecture is suitable.
  • Developers interested in models fine-tuned with efficient methods like Unsloth.