Back to leaderboard

qwen/qwen3.8-27b

LM-STUDIO MLX

qwen3_5 · 27B · 4bit

Excellent

Aug 18, 2026 · Apple M4 Max

google/gemma-4-e4b

LM-STUDIO GGUF

gemma4 · Q4_K_M

Good

Jul 7, 2026 · AMD Ryzen 7 5800H

Global Score
87 vs 68
Hardware Fit
83 vs 93
Quality Score
88 vs 57

Hardware

qwen/qwen3.8-27b google/gemma-4-e4b
MachineMacBook ProLENOVO 82JQ
CPUApple M4 MaxAMD Ryzen 7 5800H
Cores1616
RAM128 GB16 GB
GPUApple M4 MaxNVIDIA GeForce RTX 3060 Laptop GPU
OSmacOS 26.5.2Microsoft Windows 11 家庭版 中文版 10.0.26100
Archarm64x64
Power Modebalancedbalanced

Performance

qwen/qwen3.8-27b google/gemma-4-e4b
Tokens/sec29.126.2
First chunk9 ms17 ms
TTFT650 ms572 ms
Load timeN/AN/A
Memory usage21.0 GB6.5 GB
Memory %16%41%

HW Fit Score Breakdown

qwen/qwen3.8-27b

Speed
33/50
TTFT
20/20
Memory
30/30

google/gemma-4-e4b

Speed
46/50
TTFT
20/20
Memory
27/30

Quality

qwen/qwen3.8-27b

Reasoning
18/20
Coding
18/20
Instruction
16/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong

google/gemma-4-e4b

Reasoning
17/20
Coding
19/20
Instruction
4/20
Structured
4/15
Math
12/15
Multilingual
1/10
Reasoning: Strong Coding: Strong Instruction Following: Poor Structured Output: Weak Math: Strong Multilingual: Poor
View qwen/qwen3.8-27b 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