stefra/llama-NEAR-full

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:May 11, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

The stefra/llama-NEAR-full is an 8 billion parameter Llama 3.1 instruction-tuned causal language model, finetuned by stefra. This model was optimized for faster training using Unsloth and Huggingface's TRL library. It is designed for general-purpose natural language understanding and generation tasks, leveraging the Llama 3.1 architecture.

Loading preview...

Overview

The stefra/llama-NEAR-full is an 8 billion parameter instruction-tuned language model, developed by stefra. It is finetuned from the unsloth/Llama-3.1-8B-Instruct-bnb-4bit base model. A key characteristic of this model is its training methodology, which utilized Unsloth and Huggingface's TRL library to achieve a 2x faster finetuning process.

Key Capabilities

  • Instruction Following: As an instruction-tuned model, it is designed to understand and execute commands given in natural language.
  • Llama 3.1 Architecture: Benefits from the robust capabilities and performance of the Llama 3.1 model family.
  • Efficient Training: The use of Unsloth indicates an optimization for faster and potentially more resource-efficient finetuning.

Good For

  • General NLP Tasks: Suitable for a wide range of applications requiring text generation, summarization, question answering, and conversational AI.
  • Developers Seeking Llama 3.1: Offers a readily available, finetuned Llama 3.1 variant for deployment.
  • Research into Efficient Finetuning: Can serve as an example of a model trained with Unsloth for faster iteration cycles.