qwen/qwen3.6-35b-a3b
LM-STUDIO MLXqwen3_5_moe · 35B · 4bit
Jul 17, 2026 · Apple M5 Max
llama3.1:8b
OLLAMA GGUFllama · 8.0B · Q4_K_M
Jun 26, 2026 · Cortex-A78AE
Global Score
89 vs 46
Hardware Fit
97 vs 61
Quality Score
86 vs 40
Hardware
qwen/qwen3.6-35b-a3b llama3.1:8b
MachineMacBook ProNVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPUApple M5 MaxCortex-A78AE
Cores186
RAM64 GB7 GB
GPUApple M5 MaxIntegrated / Unknown
OSmacOS 27.0Ubuntu 22.04.5 LTS
Archarm64arm64
Power Modebalancedbalanced
Performance
qwen/qwen3.6-35b-a3b llama3.1:8b
Tokens/sec130.27.8
First chunk5 ms1625 ms
TTFT135 ms1.6 s
Load timeN/A21.8 s
Memory usage26.6 GB5.2 GB
Memory %42%71%
HW Fit Score Breakdown
qwen/qwen3.6-35b-a3b
Speed
50/50
TTFT
20/20
Memory
27/30
llama3.1:8b
Speed
26/50
TTFT
20/20
Memory
15/30
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
llama3.1:8b
Reasoning
12/20
Coding
10/20
Instruction
4/20
Structured
1/15
Math
4/15
Multilingual
9/10
Reasoning: Adequate Coding: Adequate Instruction Following: Poor Structured Output: Poor 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