Back to leaderboard

qwen/qwen3.6-35b-a3b

LM-STUDIO MLX

qwen3_5_moe · 35B · 4bit

Excellent

Jul 17, 2026 · Apple M5 Max

google/gemma-4-e4b

LM-STUDIO GGUF

gemma4 · Q4_K_M

Good

Jul 7, 2026 · AMD Ryzen 7 5800H

Global Score
89 vs 62
Hardware Fit
97 vs 77
Quality Score
86 vs 55

Hardware

qwen/qwen3.6-35b-a3b google/gemma-4-e4b
MachineMacBook ProLENOVO 82JQ
CPUApple M5 MaxAMD Ryzen 7 5800H
Cores1816
RAM64 GB16 GB
GPUApple M5 MaxNVIDIA GeForce RTX 3060 Laptop GPU
OSmacOS 27.0Microsoft Windows 11 家庭版 中文版 10.0.26100
Archarm64x64
Power Modebalancedbalanced

Performance

qwen/qwen3.6-35b-a3b google/gemma-4-e4b
Tokens/sec130.227.6
First chunk5 ms23 ms
TTFT135 ms480 ms
Load timeN/AN/A
Memory usage26.6 GB13.2 GB
Memory %42%83%

HW Fit Score Breakdown

qwen/qwen3.6-35b-a3b

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

google/gemma-4-e4b

Speed
48/50
TTFT
20/20
Memory
9/30

Quality

qwen/qwen3.6-35b-a3b

Reasoning
15/20
Coding
18/20
Instruction
16/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

google/gemma-4-e4b

Reasoning
18/20
Coding
17/20
Instruction
5/20
Structured
3/15
Math
11/15
Multilingual
1/10
Reasoning: Strong Coding: Strong Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Poor
View qwen/qwen3.6-35b-a3b details | View google/gemma-4-e4b 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