Back to leaderboard

microsoft/phi-4-mini-reasoning

LM-STUDIO MLX

phi3 · 3.8B · 4bit

Good

Mar 3, 2026 · Apple M4

qwen3.6:27b

OLLAMA GGUF

qwen35 · 27.8B · Q4_K_M

Not Rec.

Jun 25, 2026 · Apple M4 Pro

Global Score
62 vs 20
Hardware Fit
99 vs 34
Quality Score
46 vs 14

Hardware

microsoft/phi-4-mi… qwen3.6:27b
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… qwen3.6:27b
Tokens/sec40.310.6
First chunk289 ms1344 ms
TTFT289 ms30.0 s
Load timeN/A0.2 s
Memory usage2.0 GB18.1 GB
Memory %6%75%

HW Fit Score Breakdown

microsoft/phi-4-mini-re…

Speed
50/50
TTFT
20/20
Memory
29/30

qwen3.6:27b

Speed
22/50
TTFT
0/20
Memory
12/30

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

qwen3.6:27b

Reasoning
2/20
Coding
1/20
Instruction
2/20
Structured
1/15
Math
2/15
Multilingual
6/10
Reasoning: Poor Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Poor Multilingual: Adequate

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