eagle25/mumbai-grpo-agent

TEXT GENERATIONConcurrency Cost:1Model Size:1.5BQuant:BF16Ctx Length:32kPublished:Apr 25, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

The eagle25/mumbai-grpo-agent is a 1.5 billion parameter Qwen2.5-based instruction-tuned causal language model developed by eagle25. This model was fine-tuned using Unsloth and Huggingface's TRL library, enabling faster training. It is designed for general instruction-following tasks, leveraging its efficient training methodology.

Loading preview...

Model Overview

The eagle25/mumbai-grpo-agent is a 1.5 billion parameter instruction-tuned language model based on the Qwen2.5 architecture. Developed by eagle25, this model was fine-tuned from unsloth/qwen2.5-1.5b-instruct-unsloth-bnb-4bit.

Key Characteristics

  • Architecture: Qwen2.5-based, a causal language model.
  • Parameter Count: 1.5 billion parameters, offering a balance between performance and computational efficiency.
  • Context Length: Supports a substantial context window of 32768 tokens, allowing for processing longer inputs and maintaining conversational coherence over extended interactions.
  • Training Efficiency: Fine-tuned using Unsloth and Huggingface's TRL library, which facilitated a 2x faster training process.
  • License: Released under the Apache-2.0 license, promoting open and flexible use.

Good For

  • Instruction Following: Optimized for general instruction-following tasks, making it suitable for a wide range of applications requiring precise responses to user prompts.
  • Efficient Deployment: Its relatively smaller size (1.5B parameters) combined with efficient training methods makes it a good candidate for applications where computational resources are a consideration.