qwen3-vl:8b
OLLAMA GGUFqwen3vl · 8.8B · Q4_K_M
Apr 28, 2026 · Intel Xeon® Gold 6230
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
72 vs 90
Hardware Fit
89 vs 96
Quality Score
64 vs 88
Hardware
qwen3-vl:8b Kimi-48B:48B
MachineHP HP Z8 G4 WorkstationHP HyperX OMEN Transcend Gaming Laptop 14-ga0xxx5CD612DXML
CPUIntel Xeon® Gold 6230NVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )
Cores8020
RAM252 GB55 GB
GPUTU104GL [Quadro RTX 5000]NVIDIA RTX Spark N1X (6144-core Blackwell RTX GPU), NVIDIA NPU
OSUbuntu 24.04.4 LTSMicrosoft Windows 11 Pro 10.0.28000
Archx64arm64
Power Modelow-powerbalanced
Performance
qwen3-vl:8b Kimi-48B:48B
Tokens/sec68.262.7
First chunk285 ms397 ms
TTFT2.0 s397 ms
Load time42.4 s16.8 s
Memory usage7.1 GB25.5 GB
Memory %3%47%
HW Fit Score Breakdown
qwen3-vl:8b
Speed
50/50
TTFT
9/20
Memory
30/30
Kimi-48B:48B
Speed
50/50
TTFT
20/20
Memory
26/30
Quality
qwen3-vl:8b
Reasoning
16/20
Coding
3/20
Instruction
10/20
Structured
11/15
Math
14/15
Multilingual
10/10
Reasoning: Strong Coding: Poor Instruction Following: Adequate Structured Output: Adequate Math: Strong Multilingual: Strong
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