qwen3.5:4b
OLLAMA GGUFqwen35 · 4.7B · Q4_K_M
Aug 22, 2026 · Apple M1 Pro
nous-hermes2:latest
OLLAMA GGUFllama · 11B · Q4_0
Mar 1, 2026 · Apple M4
Global Score
69 vs 68
Hardware Fit
85 vs 87
Quality Score
62 vs 60
Hardware
qwen3.5:4b nous-hermes2:latest
MachineMacBook ProMacBook Air
CPUApple M1 ProApple M4
Cores1010
RAM16 GB32 GB
GPUApple M1 ProApple M4
OSmacOS 26.6.2macOS 26.3
Archarm64arm64
Power Modebalancedbalanced
Performance
qwen3.5:4b nous-hermes2:latest
Tokens/sec36.915.9
First chunk474 msN/A
TTFT6.9 s345 ms
Load time2.9 s0.4 s
Memory usage5.3 GB6.6 GB
Memory %33%21%
HW Fit Score Breakdown
qwen3.5:4b
Speed
50/50
TTFT
6/20
Memory
29/30
nous-hermes2:latest
Speed
27/50
TTFT
30/20
Memory
30/30
Quality
qwen3.5:4b
Reasoning
16/20
Coding
1/20
Instruction
14/20
Structured
9/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Strong Multilingual: Strong
nous-hermes2:latest
Reasoning
11/20
Coding
9/20
Instruction
12/20
Structured
15/15
Math
4/15
Multilingual
9/10
Reasoning: Adequate Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Weak 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