microsoft/phi-4-mini-reasoning

LM-STUDIO MLX

phi3 · 3.8B · 4bit

Good

Mar 3, 2026 · Apple M4

richardyoung/qwythos-9b-abliterated:Q8_0

OLLAMA GGUF

qwen35 · 9.0B · Q8_0

Good

Jun 28, 2026 · Apple M4 Pro

Global Score

62 vs 65

Hardware Fit

99 vs 96

Quality Score

46 vs 52

Hardware

microsoft/phi-4-mi… richardyoung/qwyth…

MachineMacBook AirMac mini

CPUApple M4Apple M4 Pro

Cores1012

RAM32 GB24 GB

GPUApple M4Apple M4 Pro

OSmacOS 26.3macOS 15.6.1

Archarm64arm64

Power Modebalancedbalanced

Performance

microsoft/phi-4-mi… richardyoung/qwyth…

Tokens/sec40.325.3

First chunk289 ms397 ms

TTFT289 ms397 ms

Load timeN/A6.3 s

Memory usage2.0 GB8.3 GB

Memory %6%34%

HW Fit Score Breakdown

microsoft/phi-4-mini-re…

Speed

50/50

TTFT

20/20

Memory

29/30

richardyoung/qwythos-9b…

Speed

47/50

TTFT

20/20

Memory

29/30

Quality

microsoft/phi-4-mini-re…

Reasoning

7/20

Coding

6/20

Instruction

7/20

Structured

7/15

Math

12/15

Multilingual

7/10

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

richardyoung/qwythos-9b…

Reasoning

18/20

Coding

13/20

Instruction

6/20

Structured

0/15

Math

14/15

Multilingual

1/10

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

View microsoft/phi-4-mi… details | View richardyoung/qwyth… 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