deepseek-r1:14b

qwen2 · 14.8B · Q4_K_M

Not Rec.

Mar 2, 2026 · Apple M4

granite3.1-dense:8b

granite · 8.2B · Q4_K_M

Good

Mar 1, 2026 · Apple M4

Global Score

26 vs 76

Hardware Fit

50 vs 91

Quality Score

10 vs 66

Hardware

deepseek-r1:14b granite3.1-dense:8b

MachineMacBook AirMacBook Air

CPUApple M4Apple M4

Cores1010

RAM32 GB32 GB

GPUApple M4Apple M4

OSmacOS 26.3macOS 26.3

Archarm64arm64

Power Modebalancedbalanced

Performance

deepseek-r1:14b granite3.1-dense:8b

Tokens/sec11.418.8

TTFT30.0 s308 ms

Load time1.0 s0.6 s

Memory usage16.5 GB13.0 GB

Memory %52%41%

HW Fit Score Breakdown

deepseek-r1:14b

Speed

20/40

TTFT

0/30

Memory

30/30

granite3.1-dense:8b

Speed

31/40

TTFT

30/30

Memory

30/30

Quality

deepseek-r1:14b

Reasoning

1/20

Coding

3/20

Instruction

1/20

Structured

0/15

Math

0/15

Multilingual

5/10

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

granite3.1-dense:8b

Reasoning

12/20

Coding

11/20

Instruction

15/20

Structured

14/15

Math

5/15

Multilingual

9/10

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

View deepseek-r1:14b details | View granite3.1-dense:8b details

Run yours and compare

npx metrillm@latest bench

Requires Node 20+ and Ollama or LM Studio running