google/gemma-3-4b

LM-STUDIO MLX

gemma3 · 4B · 4bit

Good

Mar 3, 2026 · Apple M4

Kimi-48B:48B

OLLAMA GGUF

kimi-linear · 49.1B · unknown

Excellent

Aug 19, 2026 · Intel Core™ Ultra X7 358H

Global Score

76 vs 86

Hardware Fit

100 vs 78

Quality Score

66 vs 89

Hardware

google/gemma-3-4b Kimi-48B:48B

MachineMacBook AirGMKtec NucBox EVO-T2S

CPUApple M4Intel Core™ Ultra X7 358H

Cores1016

RAM32 GB64 GB

GPUApple M4Intel(R) Arc(TM) B390 GPU

OSmacOS 26.3Microsoft Windows 11 Pro 10.0.26200

Archarm64x64

Power Modebalancedbalanced

Performance

google/gemma-3-4b Kimi-48B:48B

Tokens/sec39.422.8

First chunk468 ms721 ms

TTFT468 ms721 ms

Load timeN/A23.0 s

Memory usage2.8 GB25.5 GB

Memory %9%40%

HW Fit Score Breakdown

google/gemma-3-4b

Speed

50/50

TTFT

20/20

Memory

30/30

Kimi-48B:48B

Speed

31/50

TTFT

20/20

Memory

27/30

Quality

google/gemma-3-4b

Reasoning

10/20

Coding

17/20

Instruction

13/20

Structured

12/15

Math

4/15

Multilingual

10/10

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

Kimi-48B:48B

Reasoning

18/20

Coding

19/20

Instruction

15/20

Structured

15/15

Math

13/15

Multilingual

9/10

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

View google/gemma-3-4b details | View Kimi-48B:48B 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