All models

Google DeepMind’s efficient multimodal family

Gemma 4

by Google DeepMind

Chat Vision Tools Thinking

Gemma 4 pairs vision, reasoning and tool use with compact sizes (including edge-friendly E2B/E4B variants) and a 256K context window.

Publisher
Google DeepMind
Context window
256K tokens
Sizes
12B, 26B, 31B, E2B, E4B
Licence
Apache 2.0
Model card on Hugging Face

Run Gemma 4

Install it on a Pendra worker, then call it through the OpenAI-compatible API with a pdr_sk_ key.

Chat

from pendra import Pendra

client = Pendra(api_key="pdr_sk_...")

response = client.chat.completions.create(
    model="gemma4:12b",
    messages=[{"role": "user", "content": "Hello!"}],
)

print(response.choices[0].message.content)

Vision

from pendra import Pendra

client = Pendra(api_key="pdr_sk_...")

response = client.chat.completions.create(
    model="gemma4:12b",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "What's in this image?"},
                {"type": "image_url", "image_url": {"url": "https://example.com/photo.jpg"}},
            ],
        }
    ],
)

print(response.choices[0].message.content)

Run Gemma 4 on your own UK infrastructure

Deploy a worker, install Gemma 4, and start serving it through one sovereign API endpoint.