Skip to content
ShilpBot
AI Models

Google Gemma 4 26B A4B (Free): Specs, Limits and How to Use It

Google Gemma 4 26B A4B free: instruction-tuned mixture-of-experts model, image/text/video input, 262K context. Specs, limits, test prompts and how to try it.

By ShilpBot Team · Published · 3 min read

Gemma 4 26B A4B is an instruction-tuned mixture-of-experts model from Google DeepMind. Of its 25.2 billion parameters only about 3.8 billion are active for each token. The listing positions it as delivering near 31B-class quality.

Because it activates so few parameters, it is cheap to serve, which is why it has a low paid price. It is a good default for everyday tasks where you want multimodal input without paying for a large model.

Specifications checked against the public OpenRouter catalogue on 2026-11-16. See the live model page for current limits.

Google Gemma 4 26B A4B quick facts

DetailValue
DeveloperGoogle DeepMind
Model IDgoogle/gemma-4-26b-a4b-it:free
Added to the catalogue2026-04-03
TypeInstruction-tuned mixture-of-experts model
Parameters3.8B active / 25.2B total
Context window262K tokens (262,144)
Maximum output32,768 tokens
Input typesimage, text, video
PriceFree (paid listing: $0.090 in / $0.30 out per 1M tokens)

Feature support from the catalogue: Tool calling: listed; JSON / structured output: listed; Reasoning: supported, effort not adjustable.

What Google Gemma 4 26B A4B is good for

  • Everyday chat, summarisation and drafting at low cost.
  • Image and video understanding in lightweight workflows.
  • Structured output tasks, since JSON response format is listed.

Limits of Google Gemma 4 26B A4B

  • The output cap is 32,768 tokens, lower than several newer models.
  • Near-31B quality is the vendor's positioning. Compare it against the 31B model on your own tasks.
  • Free usage is rate limited and may be logged.

How Google Gemma 4 26B A4B compares

  • Google Gemma 4 31B: Both have a 262,144-token context window; both cap output at 32,768 tokens; it activates 3.8B parameters per token against 30.7B.
  • NVIDIA Nemotron 3 Nano Omni: It has a larger context window (262,144 against 256,000 tokens); its output cap is 32,768 tokens against 65,536; it does not accept audio input; it activates 3.8B parameters per token against 3B.
  • Dots Studio Dots3-Note Preview: It has a smaller context window (262,144 against 512,000 tokens); its output cap is 32,768 tokens against 460,800; it also accepts video input; it activates 3.8B parameters per token against 16B.

Calling Google Gemma 4 26B A4B from code

Use the model ID google/gemma-4-26b-a4b-it:free with any OpenAI-compatible client pointed at OpenRouter:

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "google/gemma-4-26b-a4b-it:free", "messages": [{"role": "user", "content": "Explain what you can do in two sentences."}]}'

Prompts that show what Google Gemma 4 26B A4B can do

  1. Ask it to summarise an article in three bullet points, then in one sentence, and check consistency.
  2. Upload a photo of a whiteboard and ask it to transcribe and tidy the notes.
  3. Run the same prompt on this model and on Gemma 4 31B and compare quality and speed.

What Google Gemma 4 26B A4B costs

The free listing costs nothing, but the same model also has a paid listing at about $0.090 per million input tokens and $0.30 per million output tokens. As a rough guide, sending 1 million input tokens and receiving 100,000 output tokens would cost about $0.12 on the paid listing. Use the free tier to test, and move to the paid route if you need steadier availability. Free pricing is set by the provider and can change at any time, so check the model page before you plan around it.

Keep reading

All free models are compared in our free AI chat models roundup. New to terms like mixture-of-experts? Start with how neural networks learn.

Frequently asked questions

What does A4B mean in the name?

It refers to the roughly 4 billion parameters that are active per token in this mixture-of-experts model, out of about 25 billion in total.

Is Google Gemma 4 26B A4B free, and what are its limits?

As of 2026-11-16 the catalogue lists it at zero price. 262K tokens (262,144) context window, up to 32,768 output tokens, model ID google/gemma-4-26b-a4b-it:free. Free use can be rate limited and providers may change it.

#Google Gemma 4 26B A4B#Google DeepMind free model#free AI model#AI chat model#efficient model#Kilo Gateway#OpenRouter