LLM-OLLAMA et SECURITE: microsoft

Phi-4 Model Card

Phi-4 Technical Report microsoft/phi-4 · Hugging Face

Model Summary

Developers Microsoft Research

Description phi-4 is a state-of-the-art open model built upon a blend of synthetic datasets, data from filtered public domain websites, and acquired academic books and Q&A datasets. The goal of this approach was to ensure that small capable models were trained with data focused on high quality and advanced reasoning.

phi-4 underwent a rigorous enhancement and alignment process, incorporating both supervised fine-tuning and direct preference optimization to ensure precise instruction adherence and robust safety measures

Architecture 14B parameters, dense decoder-only Transformer model

Inputs Text, best suited for prompts in the chat format

Context length 16K tokens

GPUs 1920 H100-80G

Training time 21 days

Training data 9.8T tokens

Outputs Generated text in response to input

Dates October 2024 – November 2024

Status Static model trained on an offline dataset with cutoff dates of June 2024 and earlier for publicly available data

Release date December 12, 2024

License MIT

microsoft/phi-4 · Hugging Face

https://huggingface.co/microsoft/phi-4#:~:text=phi-4%20is%20a%20state-of-the-art%20open%20model%20built%20upon,websites%2C%20and%20acquired%20academic%20books%20and%20Q%26A%20datasets.

Quick Links

Inference Providers documentation

Nebius AI Studio documentation

Compare Inference Providers for this model

microsoft/phi-4 · Hugging Face

import os

from openai import OpenAI

client = OpenAI(

base_url="https://router.huggingface.co/v1",

api_key=os.environ["HF_TOKEN"],

)

completion = client.chat.completions.create(

model="microsoft/phi-4:nebius",

messages=[

{

"role": "user",

"content": "What is the capital of France?"

}

)

print(completion.choices[0].message)

-----------------------------

Pierre Erol GIRAUDY

https://www.erolgiraudy.eu/

https://uga-ia.blogspot.com/

https://www.erolgiraudy.eu/2024/10/mes-15-livres.html

https://and500.blogspot.com/

https://www.ugaia.eu/

Pour Info : Mon livre https://amzn.eu/d/eTuHn56 sur AMAZON

Users Group Artificial Intelligence Andorra (U.G.A.I.A.) : Liste des Certificats PDF Microsoft Learn

Info SharePoint-Teams-Copilot

https://clubsp2013.blogspot.com/p/portfolio-microsoft-learn-pierre-erol.html

In this video, we dive into Phi-4 LLM, Microsoft’s latest large language model! 🚀 Get ready to: ✨ Discover the Benchmarks: We’ll review Phi-4’s impressive performance metrics—it’s setting new standards in reasoning and problem-solving! 💻 Install Phi-4 Locally: Follow our step-by-step guide to install Phi-4 LLM on your local system using Ollama for seamless setup. 🧠 Put Phi-4 to the Test: We’ll ask Phi-4 some challenging questions to see if it truly lives up to the bold claims in the technical paper about its reasoning and problem-solving capabilities. Whether you’re an AI enthusiast or just curious about Microsoft’s advancements in AI, this video has everything you need to know about Phi-4 LLM. #Phi4LLM, #MicrosoftAI, #ArtificialIntelligence, #AIResearch, #LLM, #ReasoningAI, #ProblemSolvingAI, #Ollama, #TechTutorial, #AIInnovation, #MachineLearning, #AIDevelopment, #FutureOfAI, #AITools, #TechExplained Links: Arxiv Paper: https://arxiv.org/pdf/2412.08905 Hugging Face: https://huggingface.co/microsoft/phi-4 Ollama Link: https://ollama.com/library/phi4 ------------------------------------------------ Learn More: Try Out Gloud GPUs on Novita AI (Affiliate Link): https://fas.st/t/EvuzAkeX ------------------------------------------------- CHANNEL LINKS: 🕵️‍♀️ Join my Patreon for keeping up with the updates: