Back to leaderboard

microsoft/phi-4-mini-reasoning

LM-STUDIO MLX

phi3 · 3.8B · 4bit

Good

Mar 3, 2026 · Apple M4

orcarouter/Qwen3.8-27B-Uncensored:iq4_xs

OLLAMA GGUF

qwen35 · 27.3B · IQ4_XS

Excellent

Sep 11, 2026 · Intel Core™ i5-10400

Global Score
62 vs 88
Hardware Fit
99 vs 69
Quality Score
46 vs 96

Hardware

microsoft/phi-4-mi… orcarouter/Qwen3.8…
MachineMacBook AirASUS
CPUApple M4Intel Core™ i5-10400
Cores1012
RAM32 GB16 GB
GPUApple M4Todesk Virtual Display Adapter, AskLink Display Adapter (SDR), OrayIddDriver Device, NVIDIA GeForce RTX 3080
OSmacOS 26.3Microsoft Windows 10 专业版 10.0.19045
Archarm64x64
Power Modebalancedbalanced

Performance

microsoft/phi-4-mi… orcarouter/Qwen3.8…
Tokens/sec40.339.5
First chunk289 ms220 ms
TTFT289 ms3.2 s
Load timeN/A13.7 s
Memory usage2.0 GB14.2 GB
Memory %6%89%

HW Fit Score Breakdown

microsoft/phi-4-mini-re…

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

orcarouter/Qwen3.8-27B-…

Speed
50/50
TTFT
14/20
Memory
5/30

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

orcarouter/Qwen3.8-27B-…

Reasoning
19/20
Coding
19/20
Instruction
18/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

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