Back to leaderboard

qwen3.5-2b-mlx

LM-STUDIO MLX

qwen3_5 · 2B · 4bit

Good

Mar 6, 2026 · Apple M4 Pro

qwen3-vl-4b-instruct-q4_k_m.gguf

OPENAI GGUF
Good

Apr 21, 2026 · Apple M2 Max

Global Score
65 vs 62
Hardware Fit
88 vs 100
Quality Score
55 vs 46

Hardware

qwen3.5-2b-mlx qwen3-vl-4b-instru…
MachineMac miniMac Studio
CPUApple M4 ProApple M2 Max
Cores1412
RAM64 GB96 GB
GPUApple M4 ProApple M2 Max
OSmacOS 15.7.4macOS 26.1
Archarm64arm64
Power Modebalancedbalanced

Performance

qwen3.5-2b-mlx qwen3-vl-4b-instru…
Tokens/sec48.967.8
First chunk4208 ms62 ms
TTFT4.2 s64 ms
Load timeN/A0.0 s
Memory usage1.6 GB0.0 GB
Memory %3%0%

HW Fit Score Breakdown

qwen3.5-2b-mlx

Speed
50/50
TTFT
8/20
Memory
30/30

qwen3-vl-4b-instruct-q4…

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

Quality

qwen3.5-2b-mlx

Reasoning
10/20
Coding
7/20
Instruction
13/20
Structured
14/15
Math
3/15
Multilingual
8/10
Reasoning: Adequate Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Poor Multilingual: Strong

qwen3-vl-4b-instruct-q4…

Reasoning
16/20
Coding
0/20
Instruction
12/20
Structured
1/15
Math
8/15
Multilingual
9/10
Reasoning: Strong Coding: Poor Instruction Following: Adequate Structured Output: Poor Math: Adequate Multilingual: Strong

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