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

gemma3n:latest

OLLAMA GGUF

gemma3n · 6.9B · Q4_K_M

Excellent

Mar 6, 2026 · AMD EPYC 9654P 96-Core Processor

Global Score
51 vs 81
Hardware Fit
31 vs 100
Quality Score
60 vs 73

Hardware

llama3.3:70b-instr… gemma3n:latest
MachineDell Inc. Precision Tower 7910Supermicro AS -2015HS-TNR
CPUIntel Xeon® E5-2637 v3AMD EPYC 9654P 96-Core Processor
Cores1664
RAM96 GB128 GB
GPUNVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3060GH100 [H100L 94GB], ASPEED Graphics Family, GH100 [H100L 94GB]
OSMicrosoft Windows 10 Pro 10.0.19045Ubuntu 24.04.3 LTS
Archx64x64
Power Modebalancedperformance

Performance

llama3.3:70b-instr… gemma3n:latest
Tokens/sec1.7108.7
First chunk5553 ms333 ms
TTFT5.6 s333 ms
Load time0.4 s4.5 s
Memory usage50.3 GB8.6 GB
Memory %53%7%

HW Fit Score Breakdown

llama3.3:70b-instruct-q…

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

gemma3n:latest

Speed
50/50
TTFT
20/20
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

gemma3n:latest

Reasoning
14/20
Coding
14/20
Instruction
14/20
Structured
14/15
Math
7/15
Multilingual
10/10
Reasoning: Adequate Coding: Adequate Instruction Following: Adequate Structured Output: Strong Math: Weak 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