gemma4:26b-nvfp4

nvfp4

THINKING MODEL

MacBook Pro (Apple M4 Pro)

48 GB · macOS 26.4.1

Tested on May 25, 2026 · Submitted by jc_code

Top 12% Compare

Global Score

85 /100

Not Rec.

Hardware Fit

80/100

Quality

87/100

Get this model

🦙

Ollama

ollama pull gemma4:26b-nvfp4

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: MacBook Pro
CPU: Apple M4 Pro
Cores: 14 total (10 perf + 4 eff)
Frequency: 2.4 GHz
RAM: 48 GB LPDDR5
GPU: Apple M4 Pro
OS: macOS 26.4.1
Arch: arm64
Power Mode: balanced

Performance

Tokens/sec Tokens generated per second — higher is better: 65.7
Standard deviation Variation of speed across benchmark runs — lower means more consistent: ±0.3
First chunk latency Delay before the first streamed chunk arrives from runtime: 233 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: 5.0 s
Memory usage RAM consumed during inference — percentage of total system memory: 15.2 GB (32%)
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: ~775

Score breakdown

Speed

50/50

Time to first token

0/20

Memory

30/30

Quality

Reasoning

17/20

Coding

15/20

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 0.24.0
Model format: GGUF
Hardware profile: BALANCED
Result hash: 931ba27e801c7d450421eb97c2c0fae9a550b0e74228aa7a8f9387a6b18c1681

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

Running on battery power — performance may be reduced.

Disqualifiers

Time to first token too high: 30000ms (maximum: 17841ms for BALANCED profile)

Bench Environment

Power: Battery CPU load: avg 18% (peak 28%)

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