deepseek-coder:6.7b

llama · 7B · Q4_0

Marginal

Mar 2, 2026 · Apple M4

gemma3:4b

gemma3 · 4.3B · Q4_K_M

Excellent

Mar 2, 2026 · Apple M4

Global Score

57 vs 80

Hardware Fit

98 vs 100

Quality Score

39 vs 71

Hardware

deepseek-coder:6.7b gemma3:4b

MachineMacBook AirMacBook Air

CPUApple M4Apple M4

Cores1010

RAM32 GB32 GB

GPUApple M4Apple M4

OSmacOS 26.3macOS 26.3

Archarm64arm64

Power Modebalancedbalanced

Performance

deepseek-coder:6.7b gemma3:4b

Tokens/sec24.134.6

TTFT251 ms303 ms

Load time0.9 s1.2 s

Memory usage12.6 GB4.6 GB

Memory %39%14%

HW Fit Score Breakdown

deepseek-coder:6.7b

Speed

38/40

TTFT

30/30

Memory

30/30

gemma3:4b

Speed

40/40

TTFT

30/30

Memory

30/30

Quality

deepseek-coder:6.7b

Reasoning

4/20

Coding

14/20

Instruction

9/20

Structured

6/15

Math

1/15

Multilingual

5/10

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

gemma3:4b

Reasoning

11/20

Coding

17/20

Instruction

14/20

Structured

14/15

Math

5/15

Multilingual

10/10

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

View deepseek-coder:6.7b details | View gemma3:4b details

Run yours and compare

npx metrillm@latest bench

Requires Node 20+ and Ollama or LM Studio running