qwen3.6:35b-a3b-coding-nvfp4
OLLAMA GGUFnvfp4
May 25, 2026 · Apple M4 Pro
deepseek-r1:70b
OLLAMA GGUFllama · 70.6B · Q4_K_M
Jun 27, 2026 · Cortex-X925
Global Score
76 vs 47
Hardware Fit
77 vs 51
Quality Score
76 vs 45
Hardware
qwen3.6:35b-a3b-co… deepseek-r1:70b
MachineMacBook ProNVIDIA NVIDIA_DGX_Spark
CPUApple M4 ProCortex-X925
Cores1420
RAM48 GB122 GB
GPUApple M4 ProDevice 2e12
OSmacOS 26.4.1Ubuntu 24.04.4 LTS
Archarm64arm64
Power Modebalancedperformance
Performance
qwen3.6:35b-a3b-co… deepseek-r1:70b
Tokens/sec72.74.7
First chunk147 ms557 ms
TTFT30.0 s557 ms
Load time5.7 s52.6 s
Memory usage19.6 GB49.5 GB
Memory %41%41%
HW Fit Score Breakdown
qwen3.6:35b-a3b-coding-…
Speed
50/50
TTFT
0/20
Memory
27/30
deepseek-r1:70b
Speed
4/50
TTFT
20/20
Memory
27/30
Quality
qwen3.6:35b-a3b-coding-…
Reasoning
19/20
Coding
9/20
Instruction
13/20
Structured
12/15
Math
13/15
Multilingual
10/10
Reasoning: Strong Coding: Weak Instruction Following: Adequate Structured Output: Strong Math: Strong Multilingual: Strong
deepseek-r1:70b
Reasoning
3/20
Coding
1/20
Instruction
14/20
Structured
9/15
Math
8/15
Multilingual
10/10
Reasoning: Poor Coding: Poor Instruction Following: Adequate Structured Output: Adequate 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