Back to leaderboard

qwen3.5-9b-mlx

LM-STUDIO MLX

qwen3_5 · 9B · 4bit

Good

Mar 12, 2026 · Apple M4 Pro

gemma-2-2b-it

LM-STUDIO GGUF

gemma2 · 2B · Q4_K_M

Marginal

Mar 6, 2026 · Intel Core™ i5-5300U

Global Score
77 vs 58
Hardware Fit
67 vs 50
Quality Score
81 vs 62

Hardware

qwen3.5-9b-mlx gemma-2-2b-it
MachineMac miniLENOVO 20BUS00700
CPUApple M4 ProIntel Core™ i5-5300U
Cores144
RAM64 GB16 GB
GPUApple M4 ProIntel(R) HD Graphics 5500
OSmacOS 15.7.4Microsoft Windows 10 Professionnel 10.0.19045
Archarm64x64
Power Modebalancedbalanced

Performance

qwen3.5-9b-mlx gemma-2-2b-it
Tokens/sec20.33.6
First chunk4216 ms4509 ms
TTFT4.2 s4.5 s
Load timeN/AN/A
Memory usage5.6 GB1.6 GB
Memory %9%10%

HW Fit Score Breakdown

qwen3.5-9b-mlx

Speed
29/50
TTFT
8/20
Memory
30/30

gemma-2-2b-it

Speed
14/50
TTFT
12/20
Memory
24/30

Quality

qwen3.5-9b-mlx

Reasoning
15/20
Coding
18/20
Instruction
16/20
Structured
14/15
Math
8/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong

gemma-2-2b-it

Reasoning
11/20
Coding
8/20
Instruction
14/20
Structured
15/15
Math
4/15
Multilingual
10/10
Reasoning: Adequate Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Poor 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