Back to leaderboard

qwen3.5-4b-mlx

LM-STUDIO MLX

qwen3_5 · 4B · 4bit

Not Rec.

Mar 6, 2026 · Apple M4 Pro

qwen2.5:1.5b

OLLAMA GGUF

qwen2 · 1.5B · Q4_K_M

Good

Mar 16, 2026 · Intel Core™ i7-4771

Global Score
62 vs 65
Hardware Fit
35 vs 92
Quality Score
73 vs 54

Hardware

qwen3.5-4b-mlx qwen2.5:1.5b
MachineMac miniiMac
CPUApple M4 ProIntel Core™ i7-4771
Cores148
RAM64 GB32 GB
GPUApple M4 ProNVIDIA GeForce GTX 780M
OSmacOS 15.7.4macOS 15.7.3
Archarm64x64
Power Modebalancedbalanced

Performance

qwen3.5-4b-mlx qwen2.5:1.5b
Tokens/sec0.019.6
First chunk5548 ms690 ms
TTFT5.5 s690 ms
Load timeN/A21.4 s
Memory usage2.9 GB1.0 GB
Memory %5%3%

HW Fit Score Breakdown

qwen3.5-4b-mlx

Speed
0/50
TTFT
6/20
Memory
29/30

qwen2.5:1.5b

Speed
42/50
TTFT
20/20
Memory
30/30

Quality

qwen3.5-4b-mlx

Reasoning
17/20
Coding
17/20
Instruction
9/20
Structured
8/15
Math
13/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Weak Structured Output: Adequate Math: Strong Multilingual: Strong

qwen2.5:1.5b

Reasoning
8/20
Coding
9/20
Instruction
12/20
Structured
13/15
Math
5/15
Multilingual
7/10
Reasoning: Weak Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Weak Multilingual: Adequate
View qwen3.5-4b-mlx details | View qwen2.5:1.5b 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