Back to leaderboard

qwen3.5:4b-mlx

OLLAMA GGUF

nvfp4

Marginal

Aug 24, 2026 · Apple M2

GPT-20B:20b

OLLAMA GGUF

gptoss · 20.9B · MXFP4

Good

Aug 19, 2026 · Intel Core™ Ultra X7 358H

Global Score
53 vs 79
Hardware Fit
81 vs 59
Quality Score
41 vs 88

Hardware

qwen3.5:4b-mlx GPT-20B:20b
MachineMac miniGMKtec NucBox EVO-T2S
CPUApple M2Intel Core™ Ultra X7 358H
Cores816
RAM8 GB64 GB
GPUApple M2Intel(R) Arc(TM) B390 GPU
OSmacOS 26.5.2Microsoft Windows 11 Pro 10.0.26200
Archarm64x64
Power Modebalancedbalanced

Performance

qwen3.5:4b-mlx GPT-20B:20b
Tokens/sec24.819.3
First chunk317 ms2004 ms
TTFT7.4 s10.4 s
Load time0.0 s10.4 s
Memory usage3.9 GB12.9 GB
Memory %49%20%

HW Fit Score Breakdown

qwen3.5:4b-mlx

Speed
50/50
TTFT
6/20
Memory
25/30

GPT-20B:20b

Speed
26/50
TTFT
3/20
Memory
30/30

Quality

qwen3.5:4b-mlx

Reasoning
13/20
Coding
2/20
Instruction
4/20
Structured
4/15
Math
10/15
Multilingual
8/10
Reasoning: Adequate Coding: Poor Instruction Following: Poor Structured Output: Weak Math: Adequate Multilingual: Strong

GPT-20B:20b

Reasoning
19/20
Coding
18/20
Instruction
11/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate 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