localized-ft/Qwen3-8B-school-of-reward-hacks-second-third-sft-seed4

TEXT GENERATIONPricing:Input $0.468 / Output $1.82Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 23, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The localized-ft/Qwen3-8B-school-of-reward-hacks-second-third-sft-seed4 is an 8 billion parameter Qwen3 model developed by localized-ft, fine-tuned from unsloth/Qwen3-8B. This model was trained with Unsloth and Huggingface's TRL library, achieving a 2x speedup in the training process. It features a 32768 token context length and is optimized for efficient fine-tuning.

Loading preview...

Model Overview

This model, developed by localized-ft, is an 8 billion parameter Qwen3 variant fine-tuned from the unsloth/Qwen3-8B base model. It leverages the Unsloth library in conjunction with Huggingface's TRL library, which enabled a 2x faster training speed during its development. The model maintains a substantial context length of 32768 tokens.

Key Characteristics

  • Architecture: Qwen3-8B
  • Developer: localized-ft
  • Training Efficiency: Achieved 2x faster training using Unsloth and Huggingface TRL.
  • Context Length: Supports a 32768 token context window.
  • License: Released under the Apache-2.0 license.

When to Use This Model

This model is particularly suitable for developers and researchers interested in:

  • Efficient Fine-tuning: Its development highlights the benefits of using Unsloth for accelerated training.
  • Qwen3-based Applications: Ideal for tasks requiring a robust 8B parameter Qwen3 model.
  • Long Context Tasks: The 32768 token context length makes it suitable for applications requiring extensive input or output.