Kimi-48B:48B
OLLAMA GGUFkimi-linear · 49.1B · unknown
Aug 19, 2026 · Intel Core™ Ultra X7 358H
qwen3.5-4b-mlx
LM-STUDIO MLXqwen3_5 · 4B · 4bit
Mar 6, 2026 · Apple M4 Pro
Global Score
86 vs 62
Hardware Fit
78 vs 35
Quality Score
89 vs 73
Hardware
Kimi-48B:48B qwen3.5-4b-mlx
MachineGMKtec NucBox EVO-T2SMac mini
CPUIntel Core™ Ultra X7 358HApple M4 Pro
Cores1614
RAM64 GB64 GB
GPUIntel(R) Arc(TM) B390 GPUApple M4 Pro
OSMicrosoft Windows 11 Pro 10.0.26200macOS 15.7.4
Archx64arm64
Power Modebalancedbalanced
Performance
Kimi-48B:48B qwen3.5-4b-mlx
Tokens/sec22.80.0
First chunk721 ms5548 ms
TTFT721 ms5.5 s
Load time23.0 sN/A
Memory usage25.5 GB2.9 GB
Memory %40%5%
HW Fit Score Breakdown
Kimi-48B:48B
Speed
31/50
TTFT
20/20
Memory
27/30
qwen3.5-4b-mlx
Speed
0/50
TTFT
6/20
Memory
29/30
Quality
Kimi-48B:48B
Reasoning
18/20
Coding
19/20
Instruction
15/20
Structured
15/15
Math
13/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
qwen3.5-4b-mlx
Reasoning
17/20
Coding
17/20
Instruction
9/20
Structured
8/15
Math
13/15
Multilingual
9/10
Reasoning: Strong Coding: Strong Instruction Following: Weak Structured Output: Adequate 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