gemma-3-270m-it-qat-mlx
LM-STUDIO MLXgemma3_text · 270M · 4bit
Mar 4, 2026 · Apple M4
qwen3-coder-next:latest
OLLAMA GGUFqwen3next · 79.7B · Q4_K_M
Mar 5, 2026 · AMD RYZEN AI MAX+ 395
Global Score
43 vs 82
Hardware Fit
99 vs 71
Quality Score
19 vs 87
Hardware
gemma-3-270m-it-qa… qwen3-coder-next:l…
MachineMacBook AirAZW GTR Pro
CPUApple M4AMD RYZEN AI MAX+ 395
Cores1032
RAM32 GB125 GB
GPUApple M4AMD Radeon 8060S
OSmacOS 26.3Ubuntu 24.04.4 LTS
Archarm64x64
Power Modebalancedperformance
Performance
gemma-3-270m-it-qa… qwen3-coder-next:l…
Tokens/sec344.234.5
First chunk173 ms262 ms
TTFT173 ms262 ms
Load timeN/A16.4 s
Memory usage0.3 GB56.8 GB
Memory %1%46%
HW Fit Score Breakdown
gemma-3-270m-it-qat-mlx
Speed
50/50
TTFT
20/20
Memory
29/30
qwen3-coder-next:latest
Speed
34/50
TTFT
20/20
Memory
17/30
Quality
gemma-3-270m-it-qat-mlx
Reasoning
4/20
Coding
0/20
Instruction
9/20
Structured
3/15
Math
1/15
Multilingual
2/10
Reasoning: Poor Coding: Poor Instruction Following: Weak Structured Output: Poor Math: Poor Multilingual: Poor
qwen3-coder-next:latest
Reasoning
18/20
Coding
18/20
Instruction
13/20
Structured
15/15
Math
13/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong
Run yours and compare
$
npm install -g metrillm@latest$
metrillmRequires Node 20+ and Ollama or LM Studio running
Or run without installing: npx metrillm@latest