Back to leaderboard

llama3.3:70b-instruct-q4_K_M

OLLAMA GGUF

llama · 70.6B · Q4_K_M

Not Rec.

Aug 20, 2026 · Intel Xeon® E5-2637 v3

llama3.2:3b

OLLAMA GGUF

llama · 3.2B · Q4_K_M

Good

Jun 25, 2026 · Cortex-A78AE

Global Score
51 vs 74
Hardware Fit
31 vs 99
Quality Score
60 vs 63

Hardware

llama3.3:70b-instr… llama3.2:3b
MachineDell Inc. Precision Tower 7910NVIDIA NVIDIA Jetson Orin Nano Engineering Reference Developer Kit Super
CPUIntel Xeon® E5-2637 v3Cortex-A78AE
Cores166
RAM96 GB7 GB
GPUNVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3060Integrated / Unknown
OSMicrosoft Windows 10 Pro 10.0.19045Ubuntu 22.04.5 LTS
Archx64arm64
Power Modebalancedbalanced

Performance

llama3.3:70b-instr… llama3.2:3b
Tokens/sec1.721.1
First chunk5553 ms1032 ms
TTFT5.6 s1.0 s
Load time0.4 s8.4 s
Memory usage50.3 GB2.4 GB
Memory %53%32%

HW Fit Score Breakdown

llama3.3:70b-instruct-q…

Speed
2/50
TTFT
5/20
Memory
24/30

llama3.2:3b

Speed
50/50
TTFT
20/20
Memory
29/30

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

llama3.2:3b

Reasoning
10/20
Coding
14/20
Instruction
15/20
Structured
12/15
Math
3/15
Multilingual
9/10
Reasoning: Adequate Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Poor 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