qwen3:4b

OLLAMA GGUF

qwen3 · 4.0B · Q4_K_M

Not Rec.

Jun 26, 2026 · Cortex-A78AE

gemma4:26b

OLLAMA GGUF

gemma4 · 25.8B · Q4_K_M

Not Rec.

Apr 3, 2026 · Apple M4

Global Score

29 vs 33

Hardware Fit

86 vs 50

Quality Score

4 vs 25

Hardware

qwen3:4b gemma4:26b

MachineNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit SuperMac mini

CPUCortex-A78AEApple M4

Cores610

RAM7 GB24 GB

GPUIntegrated / UnknownApple M4

OSUbuntu 22.04.5 LTSmacOS 26.3

Archarm64arm64

Power Modebalancedbalanced

Performance

qwen3:4b gemma4:26b

Tokens/sec17.920.0

First chunk1010 ms941 ms

TTFT6.4 s30.0 s

Load time9.5 s0.2 s

Memory usage3.0 GB20.3 GB

Memory %40%85%

HW Fit Score Breakdown

qwen3:4b

Speed

50/50

TTFT

8/20

Memory

28/30

gemma4:26b

Speed

42/50

TTFT

0/20

Memory

8/30

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

gemma4:26b

Reasoning

14/20

Coding

1/20

Instruction

6/20

Structured

3/15

Math

0/15

Multilingual

1/10

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

View qwen3:4b details | View gemma4:26b 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