Back to leaderboard

google/gemma-4-e4b

LM-STUDIO GGUF

gemma4 · 7.9B · Q4_K_M

Excellent

Sep 20, 2026 · AMD Ryzen 9 8940HX

qwen3.5-2b-mlx

LM-STUDIO MLX

qwen3_5 · 2B · 4bit

Good

Mar 6, 2026 · Apple M4 Pro

Global Score
83 vs 65
Hardware Fit
96 vs 88
Quality Score
78 vs 55

Hardware

google/gemma-4-e4b qwen3.5-2b-mlx
MachineASUSTeK COMPUTER INC. ROG Strix G16 G614PH_G614PHMac mini
CPUAMD Ryzen 9 8940HXApple M4 Pro
Cores3214
RAM15 GB64 GB
GPUNVIDIA GeForce RTX 5050 Laptop GPU, AMD Radeon(TM) 610MApple M4 Pro
OSMicrosoft Windows 11 Home Single Language 10.0.26200macOS 15.7.4
Archx64arm64
Power Modebalancedbalanced

Performance

google/gemma-4-e4b qwen3.5-2b-mlx
Tokens/sec35.948.9
First chunk9 ms4208 ms
TTFT182 ms4.2 s
Load timeN/AN/A
Memory usage6.5 GB1.6 GB
Memory %43%3%

HW Fit Score Breakdown

google/gemma-4-e4b

Speed
49/50
TTFT
20/20
Memory
27/30

qwen3.5-2b-mlx

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

Quality

google/gemma-4-e4b

Reasoning
14/20
Coding
18/20
Instruction
15/20
Structured
15/15
Math
6/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Weak Multilingual: Strong

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 google/gemma-4-e4b 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