Back to leaderboard

deepseek-r1:14b

qwen2 · 14.8B · Q4_K_M

Not Rec.

Mar 1, 2026 · Apple M4

phi4:14b

phi3 · 14.7B · Q4_K_M

Good

Mar 2, 2026 · Apple M4

Global Score
25 vs 75
Hardware Fit
46 vs 79
Quality Score
11 vs 73

Hardware

deepseek-r1:14b phi4:14b
MachineMacBook AirMacBook Air
CPUApple M4Apple M4
Cores1010
RAM32 GB32 GB
GPUApple M4Apple M4
OSmacOS 26.3macOS 26.3
Archarm64arm64
Power Modeunknownbalanced

Performance

deepseek-r1:14b phi4:14b
Tokens/sec10.811.0
TTFT30.0 s532 ms
Load time5.3 s0.7 s
Memory usage16.5 GB13.4 GB
Memory %52%42%

HW Fit Score Breakdown

deepseek-r1:14b

Speed
16/40
TTFT
0/30
Memory
30/30

phi4:14b

Speed
19/40
TTFT
30/30
Memory
30/30

Quality

deepseek-r1:14b

Reasoning
1/20
Coding
0/20
Instruction
2/20
Structured
1/15
Math
1/15
Multilingual
6/10
Reasoning: Poor Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Poor Multilingual: Adequate

phi4:14b

Reasoning
15/20
Coding
18/20
Instruction
10/20
Structured
13/15
Math
7/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Weak Multilingual: Strong
View deepseek-r1:14b details | View phi4:14b details

Run yours and compare

npx metrillm@latest bench

Requires Node 20+ and Ollama or LM Studio running