qwen3.5:4b
OLLAMA GGUFqwen35 · 4.7B · Q4_K_M
Aug 22, 2026 · Apple M1 Pro
qwen2.5:1.5b
OLLAMA GGUFqwen2 · 1.5B · Q4_K_M
Jun 17, 2026 · ZHAOXIN KaiXian KX-7000 KaiXian KX-7000
Global Score
69 vs 65
Hardware Fit
85 vs 87
Quality Score
62 vs 56
Hardware
qwen3.5:4b qwen2.5:1.5b
MachineMacBook ProShanghai Zhaoxin Semiconductor Co., Ltd. ZXE CRB
CPUApple M1 ProZHAOXIN KaiXian KX-7000 KaiXian KX-7000
Cores108
RAM16 GB30 GB
GPUApple M1 ProZX C-1190
OSmacOS 26.6.2Microsoft Windows 11 Pro 10.0.26100
Archarm64x64
Power Modebalancedbalanced
Performance
qwen3.5:4b qwen2.5:1.5b
Tokens/sec36.916.8
First chunk474 ms941 ms
TTFT6.9 s941 ms
Load time2.9 s0.2 s
Memory usage5.3 GB1.0 GB
Memory %33%3%
HW Fit Score Breakdown
qwen3.5:4b
Speed
50/50
TTFT
6/20
Memory
29/30
qwen2.5:1.5b
Speed
37/50
TTFT
20/20
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
qwen2.5:1.5b
Reasoning
8/20
Coding
10/20
Instruction
12/20
Structured
13/15
Math
5/15
Multilingual
8/10
Reasoning: Weak Coding: Adequate 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