Back to leaderboard

microsoft/phi-4-mini-reasoning

LM-STUDIO MLX

phi3 · 3.8B · 4bit

Good

Mar 3, 2026 · Apple M4

Qwen3.8-27B-UD-Q3_K_XL

OLLAMA GGUF

llama · 27.3B · Q4_K

Excellent

Aug 23, 2026 · Intel Gen Intel® Core™ i5-12600K

Global Score
62 vs 90
Hardware Fit
99 vs 97
Quality Score
46 vs 87

Hardware

microsoft/phi-4-mi… Qwen3.8-27B-UD-Q3_…
MachineMacBook Air
CPUApple M4Intel Gen Intel® Core™ i5-12600K
Cores1016
RAM32 GB31 GB
GPUApple M4AlderLake-S GT1, AD106 [GeForce RTX 4060 Ti 16GB]
OSmacOS 26.3Ubuntu 24.04.4 LTS
Archarm64x64
Power Modebalancedlow-power

Performance

microsoft/phi-4-mi… Qwen3.8-27B-UD-Q3_…
Tokens/sec40.339.0
First chunk289 ms305 ms
TTFT289 ms305 ms
Load timeN/A0.0 s
Memory usage2.0 GB12.5 GB
Memory %6%41%

HW Fit Score Breakdown

microsoft/phi-4-mini-re…

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

Qwen3.8-27B-UD-Q3_K_XL

Speed
50/50
TTFT
20/20
Memory
27/30

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

Qwen3.8-27B-UD-Q3_K_XL

Reasoning
17/20
Coding
19/20
Instruction
15/20
Structured
15/15
Math
11/15
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