Back to leaderboard

qwen3.5:35b-a3b

OLLAMA GGUF

qwen35moe · 36.0B · Q4_K_M

Excellent

Mar 22, 2026 · AMD RYZEN AI MAX+ 395

qwen2.5:1.5b

OLLAMA GGUF

qwen2 · 1.5B · Q4_K_M

Good

Jun 17, 2026 · ZHAOXIN KaiXian KX-7000 KaiXian KX-7000

Global Score
88 vs 65
Hardware Fit
90 vs 87
Quality Score
87 vs 56

Hardware

qwen3.5:35b-a3b qwen2.5:1.5b
MachineAZW GTR ProShanghai Zhaoxin Semiconductor Co., Ltd. ZXE CRB
CPUAMD RYZEN AI MAX+ 395ZHAOXIN KaiXian KX-7000 KaiXian KX-7000
Cores328
RAM125 GB30 GB
GPURadeon 8060SZX C-1190
OSUbuntu 24.04.4 LTSMicrosoft Windows 11 Pro 10.0.26100
Archx64x64
Power Modeperformancebalanced

Performance

qwen3.5:35b-a3b qwen2.5:1.5b
Tokens/sec42.016.8
First chunk243 ms941 ms
TTFT243 ms941 ms
Load time9.7 s0.2 s
Memory usage32.1 GB1.0 GB
Memory %26%3%

HW Fit Score Breakdown

qwen3.5:35b-a3b

Speed
40/50
TTFT
20/20
Memory
30/30

qwen2.5:1.5b

Speed
37/50
TTFT
20/20
Memory
30/30

Quality

qwen3.5:35b-a3b

Reasoning
17/20
Coding
19/20
Instruction
15/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong 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
View qwen3.5:35b-a3b details | View qwen2.5:1.5b 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