Several new AI models, including Dolphin Mixtral, Dolphin-2_6-phi-2, Phi-2-coder, and Dolphin Phi 2.6, have been introduced by @erhartford and others. These models are designed for coding tasks, chat, and instruction following. They are based on the Mixtral mixture of experts model and the 2.7B Phi model by Microsoft Research. The models are small in size and can run on a wide range of machines using @ollama and #MLX. The announcement includes the availability of weights for running the models.
Ollama countdown continues... with a new model!! ollama run dolphin-phi Meet a fine-tuned model by @erhartford - Dolphin Phi 2.6! It's an uncensored model based on the 2.7B Phi model by Microsoft Research. Learn more here: https://t.co/sfZHieMQoo https://t.co/laSH5YNO7I
🚨New Model Alert🚨 Thrilled to announce **Phi-2-coder**. MS 🪟 Phi-2 fine-tuned on codeAlpaca 🦙⌨️ for code 👩💻 instruction following https://t.co/kfPlvc2R4H Test the model on a free colab: https://t.co/kgOGEHHiI2 I will also share the weights for running it with #MLX 🍏
Dolphin 2.6 Phi-2 is a new uncensored chat model by @erhartford Since this version of Dolphin is based on the small 2.7B Phi model by Microsoft Research, it's fast and can run on a wide range of machines using @ollama. https://t.co/nhPCffkr2E
The new Dolphin-2_6-phi-2 a very small open source LLM AI model that can fit on a controller board that is entirely the size of a postage stamp. Model: https://t.co/Bda9B8S6Wv
Weekend fun with Dolphin! Have you tried @erhartford's Dolphin family of models? Meet Dolphin Mixtral (48GB+ memory required), an uncensored, fine-tuned model based on Mistral's Mixtral mixture of experts model that excels at coding tasks. ollama run dolphin-mixtral Learn… https://t.co/HDIAItnLRn