Phi-4 Multimodal Instruct

Microsoft Azure·Released 2025-02-26

Open Source

Compare

Quality Index

10.0

364th of 440

Top 83%

Price/1M

$0.00

3rd cheapest

100% below median

Top 27%

Speed

16 tok/s

Top 61%

TTFT

0.37s

Pricing

Input

$0.00

per 1M tokens

Output

$0.00

per 1M tokens

Blended

$0.00

per 1M tokens

Cheaper than 73% of models. Median price is $0.31/1M tokens.

Cost Calculator

Tokens per day1M

100K100M

Daily

$0.00

Monthly

$0.00

vs. Similar Models

Qwen2.5 Coder Instruct 7B Q:0.0

$0.00

Gemini 1.0 UltraQ:+0.1

$0.00

Gemma 3n E4B Instruct Preview (May '25)Q:+0.1

$0.00

NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)Q:+0.1

$0.30

Performance

16

tokens/sec

Faster than 39% of models

0.37

seconds

Faster than 55% of models

Market Median

46 tok/s

64% slower

Median TTFT

0.43s

14% faster

Speed Comparison

o3-pro

13 tok/s-19%

GLM-4.6V (Non-reasoning)

22 tok/s+32%

Qwen3 Coder 30B A3B Instruct

26 tok/s+60%

Benchmarks

MMLU-Pro

48.5%

GPQA Diamond

31.5%

HLE

4.4%

LiveCodeBench

13.1%

SciCode

11.0%

TerminalBench HardNot evaluated

MATH-500

69.3%

AIME

9.3%

AIME 2025Not evaluated

IFBenchNot evaluated

Long Context RecallNot evaluated

Tau2Not evaluated

Market AverageTop Score

Open Source

Quick Compare

Similar Models

Qwen2.5 Coder Instruct 7B

Alibaba

Q: 10.0$0.00/1M33K ctx

Slower: 100%

Mistral Large (Feb '24)

Mistral

Q: 9.9$6.00/1M128K ctx

Slower: 100%

NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)

NVIDIA

Q: 10.1$0.30/1M

Faster: 738%

Qwen3.5 0.8B (Non-reasoning)

Alibaba

Q: 9.9$0.00/1M

Slower: 100%

Gemini 1.0 Ultra

Google

Q: 10.1$0.00/1M

Slower: 100%

Gemma 3n E4B Instruct Preview (May '25)

Google

Q: 10.1$0.00/1M

Slower: 100%

Compare all 7 models