gpt-oss:20b
OLLAMA GGUFgptoss · 20.9B · MXFP4
May 25, 2026 · Apple M4 Pro
qwen3.5-2b-mlx
LM-STUDIO MLXqwen3_5 · 2B · 8bit
Mar 3, 2026 · Apple M4
Global Score
94 vs 70
Hardware Fit
92 vs 93
Quality Score
95 vs 60
Hardware
gpt-oss:20b qwen3.5-2b-mlx
MachineMacBook ProMacBook Air
CPUApple M4 ProApple M4
Cores1410
RAM48 GB32 GB
GPUApple M4 ProApple M4
OSmacOS 26.4.1macOS 26.3
Archarm64arm64
Power Modebalancedbalanced
Performance
gpt-oss:20b qwen3.5-2b-mlx
Tokens/sec54.532.9
First chunk318 ms3096 ms
TTFT2.9 s3.1 s
Load time4.0 sN/A
Memory usage16.4 GB2.5 GB
Memory %34%8%
HW Fit Score Breakdown
gpt-oss:20b
Speed
50/50
TTFT
13/20
Memory
29/30
qwen3.5-2b-mlx
Speed
50/50
TTFT
13/20
Memory
30/30
Quality
gpt-oss:20b
Reasoning
19/20
Coding
19/20
Instruction
17/20
Structured
15/15
Math
15/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
qwen3.5-2b-mlx
Reasoning
11/20
Coding
9/20
Instruction
14/20
Structured
14/15
Math
3/15
Multilingual
9/10
Reasoning: Adequate Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Poor 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