New Phi-3.5 Models from Microsoft, including new MoE
huggingface.co
New Phi-3.5 Models from Microsoft, including new MoE
1–4 of 4 posts
Re: New Phi-3.5 Models from Microsoft, including new MoE
#2Mini:
https://huggingface.co/microsoft/Phi-3.5-mini-instruct
Large MoE with impressive benchmarks: https://huggingface.co/microsoft/Phi-3.5-MoE-instruct
Vision: https://huggingface.co/microsoft/Phi-3.5-vision-instruct
Re: New Phi-3.5 Models from Microsoft, including new MoE
#3The Phi models always seem to do really well when it comes to benchmarks but then in real world performance they always fall way behind competing models.
Re: New Phi-3.5 Models from Microsoft, including new MoE
#4Does anyone have an idea what the output token limit is? I only see mention of the 128k token context window, but I bet the output limit is 4k tokens.