llama3.3:70b-instruct-q4_K_M
OLLAMA GGUFllama · 70.6B · Q4_K_M
Aug 20, 2026 · Intel Xeon® E5-2637 v3
GPT-20B:20b
OLLAMA GGUFgptoss · 20.9B · MXFP4
Aug 22, 2026 · NVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )
Global Score
51 vs 93
Hardware Fit
31 vs 91
Quality Score
60 vs 94
Hardware
llama3.3:70b-instr… GPT-20B:20b
MachineDell Inc. Precision Tower 7910HP HyperX OMEN Transcend Gaming Laptop 14-ga0xxx5CD612DXML
CPUIntel Xeon® E5-2637 v3NVIDIA RTX Spark N1X (6144-core GPU, 20-core ) RTX Spark N1X (6144-core GPU, 20-core )
Cores1620
RAM96 GB55 GB
GPUNVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3060NVIDIA RTX Spark N1X (6144-core Blackwell RTX GPU), NVIDIA NPU
OSMicrosoft Windows 10 Pro 10.0.19045Microsoft Windows 11 Pro 10.0.28000
Archx64arm64
Power Modebalancedbalanced
Performance
llama3.3:70b-instr… GPT-20B:20b
Tokens/sec1.757.8
First chunk5553 ms203 ms
TTFT5.6 s3.0 s
Load time0.4 s19.2 s
Memory usage50.3 GB12.0 GB
Memory %53%22%
HW Fit Score Breakdown
llama3.3:70b-instruct-q…
Speed
2/50
TTFT
5/20
Memory
24/30
GPT-20B:20b
Speed
50/50
TTFT
11/20
Memory
30/30
Quality
llama3.3:70b-instruct-q…
Reasoning
17/20
Coding
6/20
Instruction
6/20
Structured
11/15
Math
11/15
Multilingual
9/10
Reasoning: Strong Coding: Weak Instruction Following: Weak Structured Output: Adequate Math: Adequate Multilingual: Strong
GPT-20B:20b
Reasoning
19/20
Coding
17/20
Instruction
18/20
Structured
15/15
Math
15/15
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