Skip to content

Microsoft · Chat

Phi-3 Medium 128K Instruct

Microsoft 🇺🇸ChatPhi family
llminfo IQ
101.7
Market median price
$1.00 blended / 1M
Max context
128K
Listings
3

Evidence and sources

🤗 microsoft/Phi-3-medium-128k-instructllminfo IQ 101.7, ranked among chat models. From 29 benchmarks. Plenty of data behind it.How llminfo IQ is computedReleased 21 May 2024Maker HQ: US (source)
Sources: GenAI Prices data, FlowDot data, llminfo researchPrimary data from GenAI Prices from FlowDot: name from llminfo research: release date

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Trade-offs

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.

Input & output

from the model's type
Text
Text

Listings

 Microsoft Azurephi-3-medium-128k-instructSource: GenAI Prices dataNot published by the provider$1.00$1.00
 OpenRoutermicrosoft/phi-3-medium-128k-instructSource: GenAI Prices dataNot published by the provider$1.00$1.00

1 listing without a price is hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Long context

BABILong54 related benchmarks · 54 results
qa1 0k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa1, 0k
avg(qa1-5) 0k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, avg, 0k
avg(qa1-5) 1k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, avg, 1k
avg(qa1-5) 2k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, avg, 2k
avg(qa1-5) 4k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, avg, 4k
avg(qa1-5) 8k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, avg, 8k
Show 48 more resultsHide 48 results
avg(qa1-5) 16k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, avg, 16k
avg(qa1-5) 32k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, avg, 32k
avg(qa1-5) 64k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, avg, 64k
avg(qa1-5) 128k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, avg, 128k
qa1 1k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa1, 1k
qa1 2k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa1, 2k
qa1 4k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa1, 4k
qa1 8k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa1, 8k
qa1 16k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa1, 16k
qa1 32k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa1, 32k
qa1 64k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa1, 64k
qa1 128k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa1, 128k
qa2 0k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa2, 0k
qa2 1k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa2, 1k
qa2 2k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa2, 2k
qa2 4k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa2, 4k
qa2 8k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa2, 8k
qa2 16k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa2, 16k
qa2 32k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa2, 32k
qa2 64k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa2, 64k
qa2 128k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa2, 128k
qa3 0k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa3, 0k
qa3 1k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa3, 1k
qa3 2k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa3, 2k
qa3 4k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa3, 4k
qa3 8k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa3, 8k
qa3 16k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa3, 16k
qa3 32k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa3, 32k
qa3 64k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa3, 64k
qa3 128k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa3, 128k
qa4 0k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa4, 0k
qa4 1k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa4, 1k
qa4 2k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa4, 2k
qa4 4k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa4, 4k
qa4 8k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa4, 8k
qa4 16k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa4, 16k
qa4 32k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa4, 32k
qa4 64k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa4, 64k
qa4 128k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa4, 128k
qa5 0k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa5, 0k
qa5 1k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa5, 1k
qa5 2k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa5, 2k
qa5 4k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa5, 4k
qa5 8k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa5, 8k
qa5 16k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa5, 16k
qa5 32k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa5, 32k
qa5 64k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa5, 64k
qa5 128k
Source: BABILong v0 data docs
Phi-3-medium-128k-instruct, qa5, 128k

Reasoning

Coding

Knowledge

Science

General

Instruction following

Math

Other

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · RMT-team, Kuratov et al. Official v0 numeric CSV; results licence not separately stated. · Hugging Face official benchmark leaderboards; licence varies by benchmark · pfgen-bench, Preferred Networks. Apache-2.0 except LLM-generated outputs; aggregate scores only. (licence) · NVIDIA RULER, Hsieh et al. Official README numeric table; no separate results licence asserted.

Around the web

  • Phi-2: The surprising power of small language models - Microsoft Research
    based on educational value and content quality. Secondly, we use innovative techniques to scale up, …based on educational value and content quality. Secondly, we use innovative techniques to scale up, starting from our 1.3 billion parameter model, Phi-1.5, and embedding its knowledge within the 2.7 …

    microsoft.com

  • Phi Open Models - Small Language Models | Microsoft Azure
    Phi is the Microsoft family of small language models (SLMs). These models were created to …Phi is the Microsoft family of small language models (SLMs). These models were created to give developers tools to implement AI directly on a device without the need for cloud connectivity. Phi model…

    azure.microsoft.com

  • Phi-4 Technical Report - Microsoft Research
    Related Tools Phi-4 June 18, 2025 Phi-4-multimodal and Phi-4-mini, the newest models in Microsoft’s Phi …Related Tools Phi-4 June 18, 2025 Phi-4-multimodal and Phi-4-mini, the newest models in Microsoft’s Phi family of small language models (SLMs) are now available. These models are designed to empower …

    microsoft.com

  • Textbooks Are All You Need II: phi-1.5 technical report - Microsoft Research
    December 11, 2023 The language model phi-1.5 is a Transformer with 1.3 billion parameters. It …December 11, 2023 The language model phi-1.5 is a Transformer with 1.3 billion parameters. It was trained using the same data sources as phi-1, augmented with a new data source that consists of vario…

    microsoft.com

  • Phi Open Models - Small Language Models | Microsoft Azure
    AI, that went on to drastically change the tech landscape. Phi refers to the family …AI, that went on to drastically change the tech landscape. Phi refers to the family of models as a whole. Within the Phi model family, there are multiple series of models. See below for the full list…

    azure.microsoft.com

  • Phi-4-reasoning Technical Report - Microsoft Research
    Phi-4-reasoning (opens in new tab) · Phi-4-reasoning-plus (opens in new tab) AI Model Catalog Phi-4-reasoning …Phi-4-reasoning (opens in new tab) · Phi-4-reasoning-plus (opens in new tab) AI Model Catalog Phi-4-reasoning (opens in new tab) · Phi-4-reasoning-plus (opens in new tab) · Phi-4-gguf (opens in new t…

    microsoft.com

  • Introduction to Phi-3Cookbook | Microsoft Learn
    with Lee Stott, Pamela Fox, Bruno Capuano Open at Microsoft Aug 20, 2024 This session …with Lee Stott, Pamela Fox, Bruno Capuano Open at Microsoft Aug 20, 2024 This session is an overview of the Phi-3cookbook which is a collection of resources for building AI applications using Microso…

    learn.microsoft.com

  • New models added to the Phi-3 family, available on Microsoft Azure | Microsoft Azure Blog
    helps you build transformational AI experiences and The new era of compute powering Azure AI …helps you build transformational AI experiences and The new era of compute powering Azure AI solutions . At Microsoft Build 2024, we are excited to add new models to the Phi-3 family of small, open m…

    azure.microsoft.com

  • Introducing Phi-4: Microsoft’s Newest Small Language Model Specializing in Complex Reasoning | Microsoft Community Hub
    Blog Post Microsoft Foundry Blog 2 MIN READ Introducing Phi-4: Microsoft’s Newest Small Language Model …Blog Post Microsoft Foundry Blog 2 MIN READ Introducing Phi-4: Microsoft’s Newest Small Language Model Specializing in Complex Reasoning ecekamar's avatar ecekamar Icon for Microsoft rank Microsoft D…

    techcommunity.microsoft.com

  • Microsoft launches Phi-3, its smallest AI model yet | The Verge
    Skip to main content Microsoft launches Phi-3, its smallest AI model yet AI News Tech …Skip to main content Microsoft launches Phi-3, its smallest AI model yet AI News Tech Microsoft launches Phi-3, its smallest AI model yet Phi-3 learned from ‘bedtime stories’ created by other LLMs. P…

    theverge.com

Also known as: microsoft/phi-3-medium-128k-instruct

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works