microsoft/phi-4-mini-reasoning

LM-STUDIO MLX

phi3 · 3.8B · 4bit

Good

Mar 5, 2026 · Apple M4 Pro

qwen3.5:4b

OLLAMA GGUF

qwen35 · 4.7B · Q4_K_M

Good

Aug 28, 2026 · Intel Core™ i5-10210U

Global Score

64 vs 64

Hardware Fit

100 vs 69

Quality Score

48 vs 62

Hardware

microsoft/phi-4-mi… qwen3.5:4b

MachineMac miniLENOVO 20U30029MB

CPUApple M4 ProIntel Core™ i5-10210U

Cores148

RAM64 GB15 GB

GPUApple M4 ProCometLake-U GT2 [UHD Graphics]

OSmacOS 15.7.4Pop 24.04 LTS

Archarm64x64

Power Modebalancedperformance

Performance

microsoft/phi-4-mi… qwen3.5:4b

Tokens/sec90.37.4

First chunk198 ms1815 ms

TTFT198 ms1.8 s

Load timeN/A10.9 s

Memory usage2.0 GB3.5 GB

Memory %3%23%

HW Fit Score Breakdown

microsoft/phi-4-mini-re…

Speed

50/50

TTFT

20/20

Memory

30/30

qwen3.5:4b

Speed

19/50

TTFT

20/20

Memory

30/30

Quality

microsoft/phi-4-mini-re…

Reasoning

7/20

Coding

7/20

Instruction

7/20

Structured

7/15

Math

13/15

Multilingual

7/10

Reasoning: Weak Coding: Weak Instruction Following: Weak Structured Output: Weak Math: Strong Multilingual: Adequate

qwen3.5:4b

Reasoning

15/20

Coding

14/20

Instruction

6/20

Structured

12/15

Math

6/15

Multilingual

9/10

Reasoning: Adequate Coding: Adequate Instruction Following: Weak Structured Output: Strong Math: Weak Multilingual: Strong

View microsoft/phi-4-mi… details | View qwen3.5:4b details

Run yours and compare

$ npm install -g metrillm@latest

$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest