llama4:scout
OLLAMA GGUFllama4 · 108.6B · Q4_K_M
Oct 3, 2026 · Cortex-X925
deepseek-v2:16b
OLLAMA GGUFdeepseek2 · 15.7B · Q4_0
Mar 2, 2026 · Apple M4
Global Score
81 vs 69
Hardware Fit
64 vs 100
Quality Score
88 vs 55
Hardware
llama4:scout deepseek-v2:16b
MachineNVIDIA NVIDIA_DGX_SparkMacBook Air
CPUCortex-X925Apple M4
Cores2010
RAM122 GB32 GB
GPUDevice 2e12Apple M4
OSUbuntu 24.04.5 LTSmacOS 26.3
Archarm64arm64
Power Modeperformancebalanced
Performance
llama4:scout deepseek-v2:16b
Tokens/sec19.656.6
First chunk677 msN/A
TTFT677 ms239 ms
Load time13.1 s1.2 s
Memory usage63.4 GB17.7 GB
Memory %52%55%
HW Fit Score Breakdown
llama4:scout
Speed
20/50
TTFT
20/20
Memory
24/30
deepseek-v2:16b
Speed
40/50
TTFT
30/20
Memory
30/30
Quality
llama4:scout
Reasoning
17/20
Coding
19/20
Instruction
16/20
Structured
13/15
Math
13/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
deepseek-v2:16b
Reasoning
7/20
Coding
9/20
Instruction
13/20
Structured
10/15
Math
8/15
Multilingual
8/10
Reasoning: Weak Coding: Weak 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