MedGm/Llama-Locali-Qwen2.5-Coder-1.5B
Llama-Locali-Qwen2.5-Coder-1.5B is a 1.5 billion parameter causal language model developed by MedGm, built upon the Qwen2.5-Coder-1.5B-Instruct base. This model is specifically fine-tuned for coding tasks, leveraging training data that necessitates adherence to the Llama 3.1 Community License. It offers a 32768 token context length and is designed for efficient code generation and related programming applications.
Loading preview...
Llama-Locali-Qwen2.5-Coder-1.5B Overview
Llama-Locali-Qwen2.5-Coder-1.5B is a 1.5 billion parameter language model developed by MedGm, specifically designed for coding tasks. It is a merged version of a QLoRA adapter into the Qwen2.5-Coder-1.5B-Instruct base model, saved in FP16 precision. The model's training incorporates data that falls under the Llama 3.1 Community License, which dictates specific naming and attribution terms.
Key Capabilities
- Code Generation: Optimized for generating and understanding code, building on the capabilities of the Qwen2.5-Coder base.
- Efficient Deployment: Available in FP16, making it suitable for environments where memory and computational efficiency are crucial.
- High Fidelity Merge: The adapter merge process has been verified to maintain high fidelity, with minimal logit differences compared to the unmerged adapter on the base model.
Good for
- Code-centric applications: Ideal for tasks requiring code generation, completion, or understanding in a programming context.
- Resource-constrained environments: Its 1.5B parameter size and FP16 format make it suitable for deployment on hardware with limited resources.
- Developers seeking Llama 3.1 license compatibility: Users who require models compatible with the Llama 3.1 Community License terms due to their training data origins.