gemma4:12b-fast

gemma4 · 11.9B · Q4_K_M

THINKING MODEL ECO MODE

ASUS (Intel Gen Intel® Core™ i7-12700)

30 GB · Ubuntu 26.04.1 LTS

Tested on November 13, 2026 · Submitted by elhoka

Top 23% Compare

Global Score

85 /100

Not Rec.

Hardware Fit

80/100

Quality

87/100

Get this model

🦙

Ollama

ollama pull gemma4:12b-fast

View on Ollama Library

ollama.com/library/gemma4

Get it in LM Studio

Search and download models directly from the app

🤗

Find on HuggingFace

GGUF versions & quantizations

Hardware

Machine: ASUS
CPU: Intel Gen Intel® Core™ i7-12700
Cores: 20 total (8 perf + 4 eff)
Frequency: 1.51 GHz
RAM: 30 GB
GPU: Alder Lake-S GT1 [UHD Graphics 770], NVIDIA GeForce RTX 3060
OS: Ubuntu 26.04.1 LTS
Arch: x64
Power Mode: low-power

Performance

Tokens/sec Tokens generated per second — higher is better: 75.9
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±6.5
First chunk latency Delay before the first streamed chunk arrives from runtime: 169 ms
Time to first token How fast the model starts responding — lower is better: 30.0 s
Load time Time to load the model into memory before inference: 8.4 s
Memory usage RAM consumed during inference — percentage of total system memory: 1.3 GB (5%)
Total tokens Total number of tokens generated across all benchmark prompts: 1460
Thinking tokens (est.) Estimated internal reasoning tokens — models with "thinking" generate more tokens: ~745

Score breakdown

Speed

50/50

Time to first token

0/20

Memory

30/30

Quality

Reasoning

18/20

Coding

14/20

Instruction following

16/20

Structured output

15/15

Math

14/15

Multilingual

10/10

Category levels

Reasoning: Strong Coding: Adequate Instruction Following: Strong Structured Output: Strong Math: Strong Multilingual: Strong

Metadata

Spec version: 0.2.1
Runtime: Ollama 0.40.13
Model format: GGUF
Hardware profile: BALANCED
Result hash: ffcbdb92df9c6b25598cc16c3c594b4cd96310943ece80b5bb74ad33c8ad58d2

Interpretation

Hardware fit: 80/100. Overall suitability: NOT RECOMMENDED (Global 85/100). Category profile: Reasoning: Strong, Coding: Adequate, Instruction Following: Strong, Structured Output: Strong, Math: Strong, Multilingual: Strong.

Warnings

System was in low-power mode during this benchmark.

Disqualifiers

Time to first token too high: 30000ms (maximum: 17116ms for BALANCED profile)

Bench Environment

Thermal: nominal CPU load: avg 19% (peak 24%)

Run yours now

$ npm install -g metrillm@latest

$ metrillm

Requires Node 20+ and Ollama or LM Studio running

Or run without installing: npx metrillm@latest