Back to leaderboard

qwen3.5-2b-mlx

LM-STUDIO MLX

qwen3_5 · 2B · 4bit

Good

Mar 6, 2026 · Apple M4 Pro

qwen3.8:27b-q8_0

OLLAMA GGUF

qwen35 · 27.3B · Q8_0

Excellent

Aug 22, 2026 · AMD EPYC 7713 64-Core Processor

Global Score
65 vs 89
Hardware Fit
88 vs 88
Quality Score
55 vs 89

Hardware

qwen3.5-2b-mlx qwen3.8:27b-q8_0
MachineMac miniDell Inc. PowerEdge XE8545
CPUApple M4 ProAMD EPYC 7713 64-Core Processor
Cores14128
RAM64 GB1007 GB
GPUApple M4 ProGA100 [A100 SXM4 80GB], GA100 [A100 SXM4 80GB], Integrated Matrox G200eW3 Graphics Controller, GA100 [A100 SXM4 80GB], GA100 [A100 SXM4 80GB]
OSmacOS 15.7.4Rocky Linux 8.10
Archarm64x64
Power Modebalancedunknown

Performance

qwen3.5-2b-mlx qwen3.8:27b-q8_0
Tokens/sec48.940.0
First chunk4208 ms177 ms
TTFT4.2 s177 ms
Load timeN/A24.7 s
Memory usage1.6 GB51.0 GB
Memory %3%5%

HW Fit Score Breakdown

qwen3.5-2b-mlx

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

qwen3.8:27b-q8_0

Speed
38/50
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

qwen3.8:27b-q8_0

Reasoning
18/20
Coding
19/20
Instruction
16/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong
View qwen3.5-2b-mlx details | View qwen3.8:27b-q8_0 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