granite4.2:3b
OLLAMA GGUFgranite · 3.7B · Q4_K_M
Aug 29, 2026 · Apple M1 Pro
Kimi-48B:48B
OLLAMA GGUFkimi-linear · 49.1B · unknown
Aug 22, 2026 · NVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )
Global Score
59 vs 90
Hardware Fit
80 vs 96
Quality Score
50 vs 88
Hardware
granite4.2:3b Kimi-48B:48B
MachineMacBook ProHP HyperX OMEN Transcend Gaming Laptop 14-ga0xxx5CD612DXML
CPUApple M1 ProNVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )
Cores1020
RAM16 GB55 GB
GPUApple M1 ProNVIDIA RTX Spark N1X (6144-core Blackwell RTX GPU), NVIDIA NPU
OSmacOS 26.6.2Microsoft Windows 11 Pro 10.0.28000
Archarm64arm64
Power Modebalancedbalanced
Performance
granite4.2:3b Kimi-48B:48B
Tokens/sec46.362.7
First chunk167 ms397 ms
TTFT167 ms397 ms
Load time6.3 s16.8 s
Memory usage12.7 GB25.5 GB
Memory %79%47%
HW Fit Score Breakdown
granite4.2:3b
Speed
50/50
TTFT
20/20
Memory
10/30
Kimi-48B:48B
Speed
50/50
TTFT
20/20
Memory
26/30
Quality
granite4.2:3b
Reasoning
18/20
Coding
10/20
Instruction
5/20
Structured
0/15
Math
15/15
Multilingual
2/10
Reasoning: Strong Coding: Adequate Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Poor
Kimi-48B:48B
Reasoning
16/20
Coding
19/20
Instruction
16/20
Structured
15/15
Math
12/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong 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