deepseek-r1:14b

qwen2 · 14.8B · Q4_K_M

Not Rec.

Mar 2, 2026 · Apple M4

stablelm2:1.6b

stablelm · 2B · Q4_0

Marginal

Mar 2, 2026 · Apple M4

Global Score

26 vs 53

Hardware Fit

50 vs 100

Quality Score

10 vs 22

Hardware

deepseek-r1:14b stablelm2:1.6b

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 stablelm2:1.6b

Tokens/sec11.493.8

TTFT30.0 s94 ms

Load time1.0 s0.4 s

Memory usage16.5 GB1.1 GB

Memory %52%3%

HW Fit Score Breakdown

deepseek-r1:14b

Speed

20/40

TTFT

0/30

Memory

30/30

stablelm2:1.6b

Speed

40/40

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

stablelm2:1.6b

Reasoning

4/20

Coding

7/20

Instruction

6/20

Structured

1/15

Math

2/15

Multilingual

2/10

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

View deepseek-r1:14b details | View stablelm2:1.6b details

Run yours and compare

npx metrillm@latest bench

Requires Node 20+ and Ollama or LM Studio running