llama3.3:70b-instruct-q4_K_M
OLLAMA GGUFllama · 70.6B · Q4_K_M
Aug 20, 2026 · Intel Xeon® E5-2637 v3
qwen3.5:4b
OLLAMA GGUFqwen35 · 4.2B · Q4_K_M
Oct 12, 2026 · AMD Ryzen 5 5500
Global Score
51 vs 81
Hardware Fit
31 vs 98
Quality Score
60 vs 74
Hardware
llama3.3:70b-instr… qwen3.5:4b
MachineDell Inc. Precision Tower 7910Microsoft WSL
CPUIntel Xeon® E5-2637 v3AMD Ryzen 5 5500
Cores1612
RAM96 GB8 GB
GPUNVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3060Integrated / Unknown
OSMicrosoft Windows 10 Pro 10.0.19045Ubuntu 26.04.1 LTS
Archx64x64
Power Modebalancedunknown
Performance
llama3.3:70b-instr… qwen3.5:4b
Tokens/sec1.793.7
First chunk5553 ms249 ms
TTFT5.6 s249 ms
Load time0.4 s10.1 s
Memory usage50.3 GB2.9 GB
Memory %53%38%
HW Fit Score Breakdown
llama3.3:70b-instruct-q…
Speed
2/50
TTFT
5/20
Memory
24/30
qwen3.5:4b
Speed
50/50
TTFT
20/20
Memory
28/30
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
qwen3.5:4b
Reasoning
15/20
Coding
13/20
Instruction
15/20
Structured
15/15
Math
7/15
Multilingual
9/10
Reasoning: Strong Coding: Adequate Instruction Following: Strong Structured Output: Strong 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