Back to leaderboard

gemma4:e4b-mlx

OLLAMA GGUF
Not Rec.

Jul 7, 2026 · Apple M2 Pro

qwen3.5-9b-mlx

LM-STUDIO MLX

qwen3_5 · 9B · 4bit

Marginal

Mar 11, 2026 · Apple M4

Global Score
78 vs 58
Hardware Fit
72 vs 68
Quality Score
81 vs 54

Hardware

gemma4:e4b-mlx qwen3.5-9b-mlx
MachineMacBook ProMac mini
CPUApple M2 ProApple M4
Cores1010
RAM16 GB24 GB
GPUApple M2 ProApple M4
OSmacOS 26.5macOS 26.3
Archarm64arm64
Power Modebalancedbalanced

Performance

gemma4:e4b-mlx qwen3.5-9b-mlx
Tokens/sec26.916.7
First chunk222 ms23 ms
TTFT222 ms14.1 s
Load time0.1 sN/A
Memory usage15.4 GB7.8 GB
Memory %96%33%

HW Fit Score Breakdown

gemma4:e4b-mlx

Speed
50/50
TTFT
20/20
Memory
2/30

qwen3.5-9b-mlx

Speed
36/50
TTFT
3/20
Memory
29/30

Quality

gemma4:e4b-mlx

Reasoning
14/20
Coding
19/20
Instruction
15/20
Structured
15/15
Math
8/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong

qwen3.5-9b-mlx

Reasoning
10/20
Coding
16/20
Instruction
4/20
Structured
9/15
Math
10/15
Multilingual
5/10
Reasoning: Adequate Coding: Strong Instruction Following: Poor Structured Output: Adequate Math: Adequate Multilingual: Adequate

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