Dolphin
About
The Dolphin family of large language models (LLMs), developed by Cognitive Computations, is known for its open-source and uncensored characteristics 123. Unlike many traditional LLMs, which incorporate safety filters, the Dolphin models provide responses without restrictions, making them particularly useful for research and specialized applications requiring unfiltered outputs. Users are advised to implement their own alignment strategies to ensure ethical usage. These models are enhanced with instruction-based datasets, often leveraging GPT-4 and GPT-3.5 for improved coding and conversational capabilities. The Dolphin models make use of diverse base models such as Llama 1, Llama 2, and Mistral, leading to variations in size, performance, and licensing. The ongoing development of the Dolphin project signals a commitment to broadening the range and addressing existing constraints.
Models(12)
Dolphin 2.6 Mixtral 8x7B
Dolphin 2.5 Mixtral 8x7B
Dolphin 2.9.2 Qwen2 72B
Dolphin 2.9 Llama 3 8B
Dolphin 2.7 Mixtral 8x7B
Dolphin 2.9.3 Mistral 7B
Dolphin 2.1 Mistral 7B
Dolphin 2.9.1 Llama 3 8B
Dolphin 2.2.1 Mistral 7B
Dolphin 2.9.2 Qwen2 7B
Dolphin 2.9.1 Yi1.5 9B
Dolphin 2.9.2 Phi-3 Medium