unsloth-phi-4

LM-STUDIO MLX

llama · 4bit

Excellent

Mar 6, 2026 · Apple M4 Pro

qwen3-4b-qwen3.6-plus-reasoning-distilled

LM-STUDIO GGUF

qwen3 · 4B · Q4_1

Good

Apr 21, 2026 · Apple M2 Max

Global Score

82 vs 65

Hardware Fit

88 vs 96

Quality Score

80 vs 51

Hardware

unsloth-phi-4 qwen3-4b-qwen3.6-p…

MachineMac miniMac Studio

CPUApple M4 ProApple M2 Max

Cores1412

RAM64 GB96 GB

GPUApple M4 ProApple M2 Max

OSmacOS 15.7.4macOS 26.1

Archarm64arm64

Power Modebalancedbalanced

Performance

unsloth-phi-4 qwen3-4b-qwen3.6-p…

Tokens/sec27.499.6

First chunk457 ms18 ms

TTFT457 ms1.6 s

Load timeN/A4.8 s

Memory usage7.7 GB0.3 GB

Memory %12%0%

HW Fit Score Breakdown

unsloth-phi-4

Speed

38/50

TTFT

20/20

Memory

30/30

qwen3-4b-qwen3.6-plus-r…

Speed

50/50

TTFT

16/20

Memory

30/30

Quality

unsloth-phi-4

Reasoning

15/20

Coding

19/20

Instruction

14/20

Structured

13/15

Math

9/15

Multilingual

10/10

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

qwen3-4b-qwen3.6-plus-r…

Reasoning

8/20

Coding

4/20

Instruction

14/20

Structured

9/15

Math

8/15

Multilingual

8/10

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

View unsloth-phi-4 details | View qwen3-4b-qwen3.6-p… 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