Back to leaderboard

swift-1.5-iq2_xs

LM-STUDIO UNKNOWN
Excellent

Nov 14, 2026 · AMD Ryzen 5 5600X 6-Core Processor

qwen3.5-4b-mlx

LM-STUDIO MLX

qwen3_5 · 4B · 4bit

Good

Mar 3, 2026 · Apple M4

Global Score
96 vs 71
Hardware Fit
94 vs 88
Quality Score
97 vs 63

Hardware

swift-1.5-iq2_xs qwen3.5-4b-mlx
MachineGigabyte Technology Co., Ltd. B550M AORUS PROMacBook Air
CPUAMD Ryzen 5 5600X 6-Core ProcessorApple M4
Cores1210
RAM47 GB32 GB
GPUAD106 [GeForce RTX 4060 Ti]Apple M4
OSArch LinuxmacOS 26.3
Archx64arm64
Power Modelow-powerbalanced

Performance

swift-1.5-iq2_xs qwen3.5-4b-mlx
Tokens/sec60.023.1
First chunk2 ms3279 ms
TTFT2.6 s3.3 s
Load timeN/AN/A
Memory usage0.9 GB2.9 GB
Memory %2%9%

HW Fit Score Breakdown

swift-1.5-iq2_xs

Speed
50/50
TTFT
14/20
Memory
30/30

qwen3.5-4b-mlx

Speed
45/50
TTFT
13/20
Memory
30/30

Quality

swift-1.5-iq2_xs

Reasoning
20/20
Coding
19/20
Instruction
19/20
Structured
15/15
Math
15/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

qwen3.5-4b-mlx

Reasoning
15/20
Coding
16/20
Instruction
6/20
Structured
6/15
Math
11/15
Multilingual
9/10
Reasoning: Adequate Coding: Strong Instruction Following: Weak Structured Output: Weak Math: Adequate Multilingual: Strong
View swift-1.5-iq2_xs details | View qwen3.5-4b-mlx 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