marcel/phi-2-openhermes-30k
marcel/phi-2-openhermes-30k is a 3 billion parameter causal language model, converted to MLX format from Microsoft's Phi-2 base model and fine-tuned with the OpenHermes dataset. It features a 2048-token context length and demonstrates balanced performance across various reasoning and common sense benchmarks, including HellaSwag and Winogrande. This model is primarily suited for general-purpose text generation and conversational AI tasks, offering a compact yet capable solution for on-device or resource-constrained deployments.
Loading preview...
Overview
marcel/phi-2-openhermes-30k is a 3 billion parameter language model, derived from Microsoft's Phi-2 and further fine-tuned with the OpenHermes dataset. This model has been converted to the MLX format, making it suitable for efficient deployment and inference on Apple silicon. It maintains a context length of 2048 tokens, offering a balance between performance and computational efficiency.
Key Capabilities
- General Text Generation: Capable of generating coherent and contextually relevant text for a variety of prompts.
- Reasoning Tasks: Shows solid performance on reasoning benchmarks like AI2 Reasoning Challenge and MMLU.
- Common Sense Understanding: Achieves strong scores on HellaSwag and Winogrande, indicating good common sense reasoning abilities.
- MLX Compatibility: Optimized for use with the MLX framework, enabling efficient local inference.
Good for
- On-device AI applications: Its compact size and MLX optimization make it ideal for local execution on compatible hardware.
- General conversational agents: Suitable for building chatbots or interactive assistants that require broad knowledge.
- Prototyping and experimentation: A good choice for developers exploring LLM capabilities with a smaller, efficient model.
- Tasks requiring balanced performance: Offers a well-rounded performance profile across various benchmarks, making it versatile for different applications.