Google Gemma 4 26B A4B Instruct inference overview

Price per 1M tokens

$0.10 (input)
$0.05 (cache hit)
$0.30 (output)

Parameters

4B (active)
26B (total)

Context Window

262k

Release Date

Apr 2026

Google Gemma 4 26B A4B Instruct inference details

Gemma 4 26B A4B is an instruction-tuned multimodal model from Google with LoRA support for task-specific customization. It handles text and image inputs and supports function calling for agentic workflows.

Created by: 

Google

License: 

apache-2.0

Model card: 

				
					import openai
import weave

# Weave autopatches OpenAI to log LLM calls to W&B
weave.init("<team>/<project>")

client = openai.OpenAI(
    # The custom base URL points to W&B Inference
    base_url='https://api.inference.wandb.ai/v1',

    # Get your API key from https://forge.coreweave.com/authorize
    # Consider setting it in the environment as OPENAI_API_KEY instead for safety
    api_key="<your-apikey>",

    # Optional: Team and project for usage tracking
    project="<team>/<project>",
)

response = client.chat.completions.create(
    model="google/gemma-4-26B-A4B-it",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Tell me a joke."}
    ],
)

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

Google Gemma 4 26B A4B Instruct resources