NOUMENOpen app
NOUMEN

microsoft · Open weights

Phi 4 Multimodal

Added Jul 26, 2025Open weights

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

Phi 4 by Microsoft. A small multimodal model that can handle images and text.

Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.07
Output
$0.11

Reference prices as published by the model provider. API example

Specifications

Context window
128K
Max output
16.4K

Side by side

Phi 4 Multimodal compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Phi 4 MultimodalmicrosoftThis modelLlama 3.2 3b InstructMetaRemove
ProvidermicrosoftMeta
Discovery tierOpen weightsOpen weights
AddedJul 26, 2025Jan 1, 2024
Limits
Context window128K131.07KBest
Max output16.38KBest8.19K
Average output—119 tokens
Pricing, USD per 1M tokens
Input$0.07$0.0306Best
Output$0.11$0.0493Best
Performance
Request success (24h)—100.0%
Benchmark indexes
Intelligence Index5.8Best5.7
Math Index—3.3
Capabilities
PDF inputNoYes
Reasoning effort——

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • Phi 4 Multimodal
  • Llama 3.2 3b Instruct

Intelligence Index

  • Phi 4 Multimodal5.8
  • Llama 3.2 3b Instruct5.7

Phi 4 Multimodal: Better than 3% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Math Index

    Composite of the math evaluations

    • Phi 4 Multimodal: —
    • Llama 3.2 3b Instruct: 3.3

Document reasoning

  • AA-LCRv1.1

    Reasoning across long documents

    • Phi 4 Multimodal: —
    • Llama 3.2 3b Instruct: 4.3%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Phi 4 Multimodal: 5.0%
    • Llama 3.2 3b Instruct: 5.3%Top

  • GPQA Diamond

    Graduate-level science questions

    • Phi 4 Multimodal: 31.5%Top
    • Llama 3.2 3b Instruct: 25.5%

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • Phi 4 Multimodal: 48.5%Top
    • Llama 3.2 3b Instruct: 34.7%

  • IFBench

    Following precise output instructions

    • Phi 4 Multimodal: —
    • Llama 3.2 3b Instruct: 26.2%

Coding

  • LiveCodeBench

    Recent competitive programming problems

    • Phi 4 Multimodal: 13.1%Top
    • Llama 3.2 3b Instruct: 8.3%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • Phi 4 Multimodal: —
    • Llama 3.2 3b Instruct: 3.3%

  • AIME

    Competition mathematics

    • Phi 4 Multimodal: 9.3%Top
    • Llama 3.2 3b Instruct: 6.7%

  • MATH-500

    Competition math problems

    • Phi 4 Multimodal: 69.3%Top
    • Llama 3.2 3b Instruct: 48.9%

Legacy evaluations

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Phi 4 Multimodal: —
    • Llama 3.2 3b Instruct: 21.1%

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Phi 4 Multimodal: —
    • Llama 3.2 3b Instruct: 4.3%

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "phi-4-multimodal-instruct",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="phi-4-multimodal-instruct",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

microsoft

Other microsoft models, starting with the same family.