Gemma 4on your own GPUs.

Gemma 4 served from GPUs you control, behind an OpenAI-compatible endpoint. Prompts and documents stay inside your network.

Start the $4,900 pilot Compare all models

This card is the installation. Pick a size on its edge and it is re-issued.

LLM_Deploy

Card no. LD-81Issued 2026-10-27Google

Gemma 4 31B

Pilot $4,900 fixed

Sizes: 31B dense
Active params: All parameters (dense)
Context: 256K
Licence: Apache 2.0
Fits on: 1× RTX 4090 24GB at 4-bit, 1× A100 40GB at FP8
Endpoint: llm.yourcompany.internal/v1
Egress: 0
Owner: YOU

The 72-hour pilot with Gemma 4.

$4,900 fixed. Gemma 4 31B is the most downloaded of the two and fits on 1× RTX 4090 24GB at 4-bit. The others fit on the same hardware or less.

Live on your data

72 h$4,900 fixed

Gemma 4 live on your data in 72 hours. The clock starts after access and prerequisites.

Your endpoint

/v1OpenAI-compatible

OpenAI-compatible endpoint. Your applications switch by changing the base URL.

Your hardware

Gemma 4 31B1× RTX 4090 24GB at 4-bit, 1× A100 40GB at FP8
Gemma 4 26B A4B1× T4 16GB at 4-bit, 1× RTX 5090 32GB at FP8

The readout

Evalon your data

Reference evaluation on your data. Readout and production scoping.

We size the hardware and source it at cost, with no markup. Ballpark: one H100 80GB ≈ $25–30k; a 4×H100 node ≈ $180–240k; smaller models run on far less.

Same SDK. Change the base URL.

The model is one parameter. Everything else in your application stays as it is.

client = OpenAI(
    base_url="https://llm.yourcompany.internal/v1",   # inside your network
    api_key="your-key",
)
reply = client.chat.completions.create(
    model="gemma-4-31b",
    messages=[{"role": "user", "content": "Summarise this contract."}],
)