> Phi-3 is an open-weights AI model from Microsoft, released 23 Apr 2024. Index score 94 (GPT-4 = 100), rank 195 of 229 ranked models.

[Models](https://themodelindex.org/models/) /[Microsoft](https://themodelindex.org/labs/microsoft/)

# Phi-3

Microsoft

Released 23 Apr 2024

Open

Date verified

[Compare](https://themodelindex.org/compare/phi-3-vs-gpt-4o/) [Announcement](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct)

9481–106

Index score

#195 · provisional, limited data

1/ 8

Core tests taken

1 of 6 areas · plus 2 anchor tests

−13

Versus the frontier at release

best then: GPT-4 Turbo (Apr 2024)

–

Price per million tokens

no public API price found

–

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

GPT-4 reached this score 13 months earlier.

## Test results · by area · tick = best by any model

Show 2 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

28%

Humanity's Last Exam

no result

Maths

evidence

FrontierMath T1-3

no result

Coding

no evidence yet

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

no evidence yet

APEX-Agents

no result

Novel problems

no evidence yet

ARC-AGI-2

no result

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#195

of 219 · 94

Epoch Capabilities Index

overall capability from many benchmarks

#162

of 239 · 121.2

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

Not in the top three among current models on any category we track.

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Facts and sources

Released

23 Apr 2024

Announcement

[huggingface.co ↗](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct)

Release entry

Phi-3

Scored as

phi-3-medium 14B

Artificial Analysis index

–

Output speed

–

Hugging Face

not linked

## Microsoft releases · around this one

Phi-2

Open

12 DEC 2023

Phi-3

Open

23 APR 2024

index 94

Phi-4

Open

12 DEC 2024

index 112

Phi-4-reasoning

Open

30 APR 2025

---
Source: https://themodelindex.org/models/phi-3/ · The Model Index · data as of 2026-10-01