Model Overview

The kazuyamaa/alfworld-lambda-grpo-v002-hull is a 4 billion parameter Qwen3 model developed by kazuyamaa. It is a fine-tuned version of kazuyamaa/Qwen3-afworld-v001, specifically optimized for tasks within the ALFWorld environment.

Key Capabilities

• Efficient Training: This model was trained with Unsloth and Huggingface's TRL library, resulting in a 2x speed improvement during the training process.
• ALFWorld Specialization: Fine-tuned for performance in the ALFWorld environment, suggesting enhanced capabilities for embodied AI tasks and interactive simulations.

Good For

• ALFWorld Research: Ideal for researchers and developers working on tasks and experiments within the ALFWorld benchmark.
• Efficient Fine-tuning: Demonstrates the effectiveness of using tools like Unsloth for accelerating model training.