qwen/qwen3.6-35b-a3b
LM-STUDIO MLXqwen3_5_moe · 35B · 4bit
Oct 20, 2026 · Apple M1 Pro
glm-4.7-flash:latest
OLLAMA GGUFglm4moelite · 29.9B · Q4_K_M
Mar 22, 2026 · AMD RYZEN AI MAX+ 395
Global Score
84 vs 82
Hardware Fit
78 vs 94
Quality Score
86 vs 77
Hardware
qwen/qwen3.6-35b-a3b glm-4.7-flash:latest
MachineMacBook ProAZW GTR Pro
CPUApple M1 ProAMD RYZEN AI MAX+ 395
Cores1032
RAM32 GB125 GB
GPUApple M1 ProRadeon 8060S
OSmacOS 15.6.1Ubuntu 24.04.4 LTS
Archarm64x64
Power Modebalancedperformance
Performance
qwen/qwen3.6-35b-a3b glm-4.7-flash:latest
Tokens/sec49.947.6
First chunk31 ms200 ms
TTFT708 ms200 ms
Load timeN/A7.3 s
Memory usage26.6 GB37.7 GB
Memory %83%30%
HW Fit Score Breakdown
qwen/qwen3.6-35b-a3b
Speed
50/50
TTFT
20/20
Memory
8/30
glm-4.7-flash:latest
Speed
44/50
TTFT
20/20
Memory
30/30
Quality
qwen/qwen3.6-35b-a3b
Reasoning
15/20
Coding
18/20
Instruction
16/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
glm-4.7-flash:latest
Reasoning
13/20
Coding
17/20
Instruction
12/20
Structured
15/15
Math
10/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Adequate Structured Output: Strong Math: Adequate 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