Back to leaderboard

qwen3.5-4b-mlx

LM-STUDIO MLX

qwen3_5 · 4B · 4bit

Good

Mar 3, 2026 · Apple M4

llama2:13b

OLLAMA GGUF

llama · 13B · Q4_0

Marginal

Mar 2, 2026 · Apple M4

Global Score
71 vs 56
Hardware Fit
88 vs 84
Quality Score
63 vs 44

Hardware

qwen3.5-4b-mlx llama2:13b
MachineMacBook AirMacBook Air
CPUApple M4Apple M4
Cores1010
RAM32 GB32 GB
GPUApple M4Apple M4
OSmacOS 26.3macOS 26.3
Archarm64arm64
Power Modebalancedbalanced

Performance

qwen3.5-4b-mlx llama2:13b
Tokens/sec23.113.5
First chunk3279 msN/A
TTFT3.3 s415 ms
Load timeN/A0.6 s
Memory usage2.9 GB10.3 GB
Memory %9%32%

HW Fit Score Breakdown

qwen3.5-4b-mlx

Speed
45/50
TTFT
13/20
Memory
30/30

llama2:13b

Speed
24/50
TTFT
30/20
Memory
30/30

Quality

qwen3.5-4b-mlx

Reasoning
15/20
Coding
16/20
Instruction
6/20
Structured
6/15
Math
11/15
Multilingual
9/10
Reasoning: Adequate Coding: Strong Instruction Following: Weak Structured Output: Weak Math: Adequate Multilingual: Strong

llama2:13b

Reasoning
9/20
Coding
3/20
Instruction
13/20
Structured
7/15
Math
3/15
Multilingual
9/10
Reasoning: Weak Coding: Poor Instruction Following: Adequate Structured Output: Weak Math: Poor Multilingual: Strong
View qwen3.5-4b-mlx details | View llama2:13b details

Run yours and compare

$ npm install -g metrillm@latest
$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest