Back to leaderboard

hf.co/TheDrummer/Cydonia-24B-v4.3-GGUF:Q4_K_M

OLLAMA GGUF

llama · 23.6B · unknown

Not Rec.

Oct 2, 2026 · Cortex-X925

qwen3.5-2b-mlx

LM-STUDIO MLX

qwen3_5 · 2B · 4bit

Good

Mar 6, 2026 · Apple M4 Pro

Global Score
30 vs 65
Hardware Fit
48 vs 88
Quality Score
22 vs 55

Hardware

hf.co/TheDrummer/C… qwen3.5-2b-mlx
MachineNVIDIA NVIDIA_DGX_SparkMac mini
CPUCortex-X925Apple M4 Pro
Cores2014
RAM122 GB64 GB
GPUDevice 2e12Apple M4 Pro
OSUbuntu 24.04.5 LTSmacOS 15.7.4
Archarm64arm64
Power Modeperformancebalanced

Performance

hf.co/TheDrummer/C… qwen3.5-2b-mlx
Tokens/sec14.448.9
First chunk5450 ms4208 ms
TTFT5.5 s4.2 s
Load time22.7 sN/A
Memory usage18.5 GB1.6 GB
Memory %15%3%

HW Fit Score Breakdown

hf.co/TheDrummer/Cydoni…

Speed
14/50
TTFT
4/20
Memory
30/30

qwen3.5-2b-mlx

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

Quality

hf.co/TheDrummer/Cydoni…

Reasoning
16/20
Coding
0/20
Instruction
0/20
Structured
0/15
Math
6/15
Multilingual
0/10
Reasoning: Strong Coding: Poor Instruction Following: Poor Structured Output: Poor Math: Weak Multilingual: Poor

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
View hf.co/TheDrummer/C… details | View qwen3.5-2b-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