Back to leaderboard

google/gemma-3-27b

LM-STUDIO MLX

gemma3 · 27B · 4bit

Not Rec.

Mar 19, 2026 · Apple M4

qwen3-vl:2b

OLLAMA GGUF

qwen3vl · 2.1B · Q4_K_M

Good

Mar 5, 2026 · AMD RYZEN AI MAX+ 395

Global Score
71 vs 62
Hardware Fit
35 vs 99
Quality Score
86 vs 46

Hardware

google/gemma-3-27b qwen3-vl:2b
MachineMac miniAZW GTR Pro
CPUApple M4AMD RYZEN AI MAX+ 395
Cores1032
RAM24 GB125 GB
GPUApple M4AMD Radeon 8060S
OSmacOS 26.3Ubuntu 24.04.4 LTS
Archarm64x64
Power Modebalancedlow-power

Performance

google/gemma-3-27b qwen3-vl:2b
Tokens/sec5.9127.4
First chunk19 ms101 ms
TTFT1.8 s630 ms
Load timeN/A5.9 s
Memory usage22.0 GB33.2 GB
Memory %92%27%

HW Fit Score Breakdown

google/gemma-3-27b

Speed
12/50
TTFT
19/20
Memory
4/30

qwen3-vl:2b

Speed
50/50
TTFT
19/20
Memory
30/30

Quality

google/gemma-3-27b

Reasoning
16/20
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

qwen3-vl:2b

Reasoning
7/20
Coding
0/20
Instruction
11/20
Structured
7/15
Math
13/15
Multilingual
8/10
Reasoning: Weak Coding: Poor Instruction Following: Adequate Structured Output: Weak Math: Strong Multilingual: Strong
View google/gemma-3-27b details | View qwen3-vl:2b 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