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

google/gemma-4-e4b

LM-STUDIO GGUF

gemma4 · Q4_K_M

Good

Jul 7, 2026 · AMD Ryzen 7 5800H

Global Score
51 vs 62
Hardware Fit
31 vs 77
Quality Score
60 vs 55

Hardware

llama3.3:70b-instr… google/gemma-4-e4b
MachineDell Inc. Precision Tower 7910LENOVO 82JQ
CPUIntel Xeon® E5-2637 v3AMD Ryzen 7 5800H
Cores1616
RAM96 GB16 GB
GPUNVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3060NVIDIA GeForce RTX 3060 Laptop GPU
OSMicrosoft Windows 10 Pro 10.0.19045Microsoft Windows 11 家庭版 中文版 10.0.26100
Archx64x64
Power Modebalancedbalanced

Performance

llama3.3:70b-instr… google/gemma-4-e4b
Tokens/sec1.727.6
First chunk5553 ms23 ms
TTFT5.6 s480 ms
Load time0.4 sN/A
Memory usage50.3 GB13.2 GB
Memory %53%83%

HW Fit Score Breakdown

llama3.3:70b-instruct-q…

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

google/gemma-4-e4b

Speed
48/50
TTFT
20/20
Memory
9/30

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

google/gemma-4-e4b

Reasoning
18/20
Coding
17/20
Instruction
5/20
Structured
3/15
Math
11/15
Multilingual
1/10
Reasoning: Strong Coding: Strong Instruction Following: Weak Structured Output: Poor Math: Strong Multilingual: Poor
View llama3.3:70b-instr… details | View google/gemma-4-e4b details

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