Qwen3-80B:80B
OLLAMA GGUFqwen3next · 79.7B · Q4_K_M
Aug 22, 2026 · NVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )
phi4-mini:latest
OLLAMA GGUFphi3 · 3.8B · Q4_K_M
Mar 2, 2026 · Apple M4
Global Score
87 vs 71
Hardware Fit
76 vs 100
Quality Score
91 vs 58
Hardware
Qwen3-80B:80B phi4-mini:latest
MachineHP HyperX OMEN Transcend Gaming Laptop 14-ga0xxx5CD612DXMLMacBook Air
CPUNVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )Apple M4
Cores2010
RAM55 GB32 GB
GPUNVIDIA RTX Spark N1X (6144-core Blackwell RTX GPU), NVIDIA NPUApple M4
OSMicrosoft Windows 11 Pro 10.0.28000macOS 26.3
Archarm64arm64
Power Modebalancedbalanced
Performance
Qwen3-80B:80B phi4-mini:latest
Tokens/sec43.636.2
First chunk409 msN/A
TTFT409 ms190 ms
Load time80.3 s0.8 s
Memory usage48.0 GB8.3 GB
Memory %88%26%
HW Fit Score Breakdown
Qwen3-80B:80B
Speed
50/50
TTFT
20/20
Memory
6/30
phi4-mini:latest
Speed
40/50
TTFT
30/20
Memory
30/30
Quality
Qwen3-80B:80B
Reasoning
18/20
Coding
19/20
Instruction
16/20
Structured
15/15
Math
13/15
Multilingual
10/10
Reasoning: Strong Coding: Strong Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong
phi4-mini:latest
Reasoning
12/20
Coding
10/20
Instruction
10/20
Structured
11/15
Math
5/15
Multilingual
10/10
Reasoning: Adequate Coding: Weak Instruction Following: Adequate Structured Output: Adequate Math: Weak 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