Back to leaderboard

Qwen3-30B:30b

OLLAMA GGUF

qwen3moe · 30.5B · Q4_K_M

Excellent

Aug 22, 2026 · NVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )

lfm2:24b-a2b

OLLAMA GGUF

lfm2moe · 23.8B · Q4_K_M

Good

May 24, 2026 · Intel Gen Intel® Core™ i7-1355U

Global Score
91 vs 78
Hardware Fit
98 vs 67
Quality Score
88 vs 82

Hardware

Qwen3-30B:30b lfm2:24b-a2b
MachineHP HyperX OMEN Transcend Gaming Laptop 14-ga0xxx5CD612DXMLDell Inc. Inspiron 15 3530
CPUNVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )Intel Gen Intel® Core™ i7-1355U
Cores2012
RAM55 GB31 GB
GPUNVIDIA RTX Spark N1X (6144-core Blackwell RTX GPU), NVIDIA NPURaptor Lake-P [Iris Xe Graphics]
OSMicrosoft Windows 11 Pro 10.0.28000cachyos rolling
Archarm64x64
Power Modebalancedlow-power

Performance

Qwen3-30B:30b lfm2:24b-a2b
Tokens/sec80.511.5
First chunk158 ms1727 ms
TTFT158 ms1.7 s
Load time21.8 s6.8 s
Memory usage20.3 GB13.9 GB
Memory %37%45%

HW Fit Score Breakdown

Qwen3-30B:30b

Speed
50/50
TTFT
20/20
Memory
28/30

lfm2:24b-a2b

Speed
22/50
TTFT
19/20
Memory
26/30

Quality

Qwen3-30B:30b

Reasoning
16/20
Coding
19/20
Instruction
17/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong

lfm2:24b-a2b

Reasoning
13/20
Coding
17/20
Instruction
16/20
Structured
15/15
Math
11/15
Multilingual
10/10
Reasoning: Adequate Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Adequate Multilingual: Strong

Run yours and compare

$ npm install -g metrillm@latest
$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest