Back to leaderboard

qwen3.5-2b-mlx

LM-STUDIO MLX

qwen3_5 · 2B · 4bit

Good

Mar 6, 2026 · Apple M4 Pro

qwen3.6:35b-a3b-q8_0

OLLAMA GGUF

qwen35moe · 36.0B · Q8_0

Excellent

Jul 7, 2026 · Cortex-X925

Global Score
65 vs 91
Hardware Fit
88 vs 99
Quality Score
55 vs 87

Hardware

qwen3.5-2b-mlx qwen3.6:35b-a3b-q8_0
MachineMac miniASUSTeK COMPUTER INC. GX10
CPUApple M4 ProCortex-X925
Cores1420
RAM64 GB122 GB
GPUApple M4 ProDevice 2e12
OSmacOS 15.7.4Ubuntu 24.04.4 LTS
Archarm64arm64
Power Modebalancedperformance

Performance

qwen3.5-2b-mlx qwen3.6:35b-a3b-q8_0
Tokens/sec48.956.2
First chunk4208 ms531 ms
TTFT4.2 s531 ms
Load timeN/A32.8 s
Memory usage1.6 GB41.0 GB
Memory %3%34%

HW Fit Score Breakdown

qwen3.5-2b-mlx

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

qwen3.6:35b-a3b-q8_0

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

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

qwen3.6:35b-a3b-q8_0

Reasoning
16/20
Coding
18/20
Instruction
16/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong 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