qwen3:4b

OLLAMA GGUF

qwen3 · 4.0B · Q4_K_M

Not Rec.

Jun 26, 2026 · Cortex-A78AE

qwen3.8-27b-mlx

LM-STUDIO MLX

qwen3_5 · 27B · 8bit

Marginal

Aug 17, 2026 · Apple M5 Max

Global Score

29 vs 59

Hardware Fit

86 vs 63

Quality Score

4 vs 58

Hardware

qwen3:4b qwen3.8-27b-mlx

MachineNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit SuperMacBook Pro

CPUCortex-A78AEApple M5 Max

Cores618

RAM7 GB64 GB

GPUIntegrated / UnknownApple M5 Max

OSUbuntu 22.04.5 LTSmacOS 27.0

Archarm64arm64

Power Modebalancedbalanced

Performance

qwen3:4b qwen3.8-27b-mlx

Tokens/sec17.918.3

First chunk1010 ms4 ms

TTFT6.4 s530 ms

Load time9.5 sN/A

Memory usage3.0 GB38.5 GB

Memory %40%60%

HW Fit Score Breakdown

qwen3:4b

Speed

50/50

TTFT

8/20

Memory

28/30

qwen3.8-27b-mlx

Speed

23/50

TTFT

20/20

Memory

20/30

Quality

qwen3:4b

Reasoning

4/20

Coding

0/20

Instruction

0/20

Structured

0/15

Math

0/15

Multilingual

0/10

Reasoning: Poor Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Poor Multilingual: Poor

qwen3.8-27b-mlx

Reasoning

19/20

Coding

14/20

Instruction

6/20

Structured

1/15

Math

13/15

Multilingual

5/10

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

View qwen3:4b details | View qwen3.8-27b-mlx 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