Skip to main content
Watch: I Built an AI App Without Leaving Databricks

Google - Mixture of experts - verified

Run Gemma 4 26B A4B locally

Gemma 4 26B A4B is a 25.8B mixture-of-experts (3.8B active) model from Google with a 262K-token context window, released under Apache 2.0 (commercial use is allowed); the smallest listed download is 14.4 GB, the Q4_0 (QAT) GGUF build from google/gemma-4-26B-A4B-it-qat-q4_0-gguf.

Official weights: google/gemma-4-26B-A4B-it - modalities: text, image, video

Total params
25.8B
Active params
3.8B
Context
262K
Smallest download
14.4 GB

Fits in

Smallest download at most 85% of memory

Downloads

The official weights and every quantization listed for this model, with file sizes from Hugging Face.

Gemma 4 26B A4B downloads and file sizes
DownloadFormatSizeRepo
Official weightsBF1651.6 GBgoogle/gemma-4-26B-A4B-it
Q4_0 (QAT)Smallest listedGGUF14.4 GBgoogle/gemma-4-26B-A4B-it-qat-q4_0-gguf
UD-Q4_K_MGGUF16.9 GBunsloth/gemma-4-26B-A4B-it-GGUF
Q8_0GGUF26.9 GBunsloth/gemma-4-26B-A4B-it-GGUF
4bitMLX15.3 GBmlx-community/gemma-4-26b-a4b-it-4bit

GB is decimal: 1 GB = 10^9 bytes. Add memory for the context and the operating system on top of the file size.

Run it

Copied verbatim from the page linked under each command. Server commands from model cards often assume a multi-GPU node; check the flags before you run them.

Ollama

Terminal
ollama run gemma4:26b

Source: ollama.com/library/gemma4

Transformers

Terminal
pip install -U transformers torch accelerate

Source: huggingface.co/google/gemma-4-26B-A4B-it

Not sure which runtime to use? Compare local LLM runtimes by platform, API and features.

License

License
Apache 2.0apache-2.0
Commercial use
Commercial use: YesCommercial use is allowed.
Gated download
No. The files download directly.

Worth knowing

Mixture-of-experts with 8 of 128 experts plus 1 shared expert active per token, so it runs close to a 4B dense model in speed while needing memory for all 25.8B parameters.

Released (per the announcement).

FAQ

How big is the Gemma 4 26B A4B download?

The official BF16 weights in google/gemma-4-26B-A4B-it are 51.6 GB. The smallest listed download is 14.4 GB, the Q4_0 (QAT) GGUF build from google/gemma-4-26B-A4B-it-qat-q4_0-gguf. Plan for more memory than the file size: the context (KV cache) and the operating system need room too.

Can I use Gemma 4 26B A4B commercially?

Gemma 4 26B A4B is released under Apache 2.0, and commercial use is allowed. The license text is at https://ai.google.dev/gemma/docs/gemma_4_license.

How much memory do I need for Gemma 4 26B A4B?

By this directory's rule of thumb it fits in 24 GB and up: the smallest listed download, 14.4 GB, is at most 85% of 24 GB, which leaves room for the context and the operating system.

How do I run Gemma 4 26B A4B with Ollama?

Run "ollama run gemma4:26b", as shown at https://ollama.com/library/gemma4. The Ollama library lists the gemma4:26b tag at 16GB - 19GB.

What is the context length of Gemma 4 26B A4B?

262,144 tokens. A longer context needs more memory for the KV cache, so you can run with a shorter one when memory is tight.

Sources

Last verified from the pages below.

Newsletter

New open weights, every week

One email per week. New model releases, open-weight drops and what they mean for what you build.

Keep exploring

Best local AI models

Our curated picks by hardware tier, with benchmarks and the reasoning.

Model API pricing and specs

First-party prices per 1M tokens when you would rather call an API.

AI benchmark leaderboards

Sourced scores, for when you need to compare quality.

Setup guides

Point your coding tools at the model you picked.