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

gemma4:e2b

OLLAMA GGUF

gemma4 · 5.1B · Q4_K_M

Good

Aug 28, 2026 · Intel Core™ i5-10210U

Global Score
51 vs 76
Hardware Fit
31 vs 82
Quality Score
60 vs 73

Hardware

llama3.3:70b-instr… gemma4:e2b
MachineDell Inc. Precision Tower 7910LENOVO 20U30029MB
CPUIntel Xeon® E5-2637 v3Intel Core™ i5-10210U
Cores168
RAM96 GB15 GB
GPUNVIDIA GeForce RTX 3090, NVIDIA GeForce RTX 3060CometLake-U GT2 [UHD Graphics]
OSMicrosoft Windows 10 Pro 10.0.19045Pop 24.04 LTS
Archx64x64
Power Modebalancedperformance

Performance

llama3.3:70b-instr… gemma4:e2b
Tokens/sec1.713.5
First chunk5553 ms1270 ms
TTFT5.6 s1.3 s
Load time0.4 s13.4 s
Memory usage50.3 GB6.5 GB
Memory %53%42%

HW Fit Score Breakdown

llama3.3:70b-instruct-q…

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

gemma4:e2b

Speed
35/50
TTFT
20/20
Memory
27/30

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

gemma4:e2b

Reasoning
10/20
Coding
15/20
Instruction
16/20
Structured
15/15
Math
7/15
Multilingual
10/10
Reasoning: Adequate Coding: Adequate Instruction Following: Strong 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