# microsoft/PhiCookBook

This is a Phi Family of SLMs book for getting started with Phi Models. Phi a family of open sourced AI models developed by Microsoft. Phi models are the most capable and cost-effective small language models (SLMs) available, outperforming models of the same size and next size up across a variety of language, reasoning, coding, and math benchmarks

Repository: https://github.com/microsoft/PhiCookBook
Canonical: https://ross.abutalabs.com/products/phicookbook
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: phi3, phi3-testing, phi3-vision, phi4, cookbook, language-model, phi-4, slm, small-language-model, phi-4-mini, phi-4-multimodal, phi4-mini, phi4-multimodal
Last push: 2026-08-26T19:52:55+00:00

## Health v2 (maintenance only)
Score: 69/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 60
- inputs: {"age_days": 848, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3795, forks 514 (observed 2026-08-28T04:08:18.965368+00:00)

## What it is
A cookbook of hands-on Jupyter notebook examples and guides for Microsoft's Phi family of small language models (SLMs), covering Phi-3 and Phi-4 variants including multimodal and mini editions. It shows how to run, fine-tune, quantize, and deploy Phi models across cloud and edge devices.

## Use cases
- get started with microsoft phi small language models
- run phi-4 on an edge device with limited compute
- fine-tune phi-3 for a custom task
- quantize phi models for on-device inference
- learn to build generative ai apps with slms
- run phi multimodal models on images and audio
- deploy phi models to cloud or local hardware

## When to choose
- you want practical notebook-driven examples for Phi models specifically
- you need to deploy capable small models on constrained hardware or edge devices
- you want official, maintained guidance covering the Phi-3 and Phi-4 families including multimodal variants

## When to avoid
- you need a production serving framework rather than learning material
- you are working with large frontier models like GPT or Llama-70B instead of SLMs
- you want a general LLM course not focused on the Phi model family

## Facets
- artifact type: learning-resource
- maturity: active
- function: llm-inference, machine-learning, developer-tools
- domain: large-language-models, artificial-intelligence, tutorials, developer-tools
- platform: python, cross-platform, cloud
- tags: phi-models, small-language-models, cookbook, jupyter-notebooks, microsoft, on-device-ai, fine-tuning, quantization

## Member repositories
- microsoft/PhiCookBook (main) score 69

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:18.965368+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:26:57.147775+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/PhiCookBook (fetched 2026-08-28T04:08:18.965368+00:00, sha 18aac0cab36a)
- Data as of 2026-08-30T08:39:29.467469+00:00.