Back to leaderboard

deepseek-coder:6.7b

llama · 7B · Q4_0

Good

Mar 2, 2026 · Apple M4

phi3:3.8b

phi3 · 3.8B · Q4_0

Good

Mar 2, 2026 · Apple M4

Global Score
63 vs 69
Hardware Fit
98 vs 100
Quality Score
39 vs 48

Hardware

deepseek-coder:6.7b phi3:3.8b
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 phi3:3.8b
Tokens/sec24.142.0
TTFT251 ms142 ms
Load time0.9 s1.1 s
Memory usage12.6 GB16.0 GB
Memory %39%50%

HW Fit Score Breakdown

deepseek-coder:6.7b

Speed
38/40
TTFT
30/30
Memory
30/30

phi3:3.8b

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

phi3:3.8b

Reasoning
12/20
Coding
3/20
Instruction
11/20
Structured
10/15
Math
5/15
Multilingual
7/10
Reasoning: Adequate Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Weak Multilingual: Adequate
View deepseek-coder:6.7b details | View phi3:3.8b details

Run yours and compare

npx metrillm@latest bench

Requires Node 20+ and Ollama or LM Studio running