Ctrl K
Back to models

Microsoft

Phi-4 Multimodal Instruct

  • Released
  • Open weights
  • MIT
Compare
Intelligence Index

Standard setting

5.8
Estimated · Artificial Analysis v4.3.2
Context window
128K tok

Performance profile

ReasoningRank #104 of 107
31.5%
GPQA Diamond

Capabilities

Context window

Modalities

Measured setting

Reasoning trade-offs

Source (opens in a new tab)
Intelligence Index v4.3.2
StandardEstimated
Intelligence Index
5.8

Pricing & access

No API offering listed in our tracked sources.

Run locally

Original download
11.1 GB
App memory
Unverified
Documented software
vLLMSetup guide (opens in a new tab)
Find GGUF downloads (opens in a new tab)Original weights (opens in a new tab)

Benchmarks

Full leaderboardSource (opens in a new tab)

Humanity’s Last Exam

Frontier reasoning

Phi-4 Multimodal Instruct · Standard#110 of 126

  1. 1Claude Opus 5.5Max61.4%
  2. 2Claude Fable 5.1Max59.1%
  3. 3Gemini 4 ArgonHigh57.1%
  5. 106K2 Horizon 0.9BReasoning5.4%
  6. 108Tiny Aya GlobalStandard5.2%
  7. 109Granite 4.0 MicroStandard5.0%
  8. 110Phi-4 Multimodal InstructStandard5.0%
  9. 111Llama 4 MaverickStandard4.9%
  10. 112NVIDIA Nemotron Nano 9B V2Reasoning4.9%
  11. 113Nemotron 3 Nano Omni 30B A3B ReasoningReasoning4.8%

Peers use their best measured result on this benchmark.