Gemma 4 26B A4B

Gemma 4 26B A4B is listed on the current BatchIn Model API surface.

Model detailAvailableMultimodal Transformer

Params

26B

Context

256K

Max Output

64K

License

Proprietary

API Surface

/v1/responses multimodal

Access

Available

Why pick it

Availability is synced from the live service catalog.
Pricing appears only when verified by the live service catalog.

Pricing

TierPublicCachedNote

Realtime$0.000 / $0.000$0.000Current published price

Batch$0.000 / $0.000$0.000Published batch rates appear here when available

Pricing is synced from the live BatchIn catalog.

Quick start

Works with common OpenAI-style integrations. Point your client to the BatchIn base URL and start testing.

Try now Open pricing

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.batchin.tech/v1",
    api_key="BATCHIN_API_KEY"
)

resp = client.responses.create(
    model="gemma-4-26b-a4b-it",
    input=[{
        "role": "user",
        "content": [
            {"type": "input_text", "text": "Describe the uploaded image"},
            {"type": "input_image", "image_url": "https://example.com/sample.png"}
        ]
    }]
)

print(resp.output_text)

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.batchin.tech/v1",
  apiKey: process.env.BATCHIN_API_KEY,
});

const resp = await client.responses.create({
  model: "gemma-4-26b-a4b-it",
  input: [{
    role: "user",
    content: [
      { type: "input_text", text: "Describe the uploaded image" },
      { type: "input_image", image_url: "https://example.com/sample.png" }
    ],
  }],
});

console.log(resp.output_text);

cURL

curl https://api.batchin.tech/v1/responses \  -H "Authorization: Bearer ***" \  -H "Content-Type: application/json" \  -d '{
    "model": "gemma-4-26b-a4b-it",
    "input": [{
      "role": "user",
      "content": [
        {"type":"input_text","text":"Describe the uploaded image"},
        {"type":"input_image","image_url":"https://example.com/sample.png"}
      ]
    }]
  }'

Specs

Architecture

Multimodal Transformer

Vendor

Google

Context window

256K

Max output

64K

Best for

google

free