gemma-4-26B-A4B-it
The memory it needs, what you may do with it, and the hardware it suits.

gemma-4-26B-A4B-it is a vision-language model: it reads images (scanned pages, photos, charts and screenshots) as well as text, published by Google on Hugging Face on 11 March 2026. It has about 25.8 billion parameters and a configured context window of 262,144 tokens. It is a mixture-of-experts model with about 4B parameters active per token.

Its card lists the licence as Apache 2.0. In short, the licence allows commercial use, modification and redistribution, with attribution and a copy of the licence, and includes a patent grant.

Hugging Face counted about 12,473,185 downloads of it in the 30 days before we checked (7 Oct 2026), and 1,606 people have liked it there.

It is built on google/gemma-4-26B-A4B, according to its card.

We size it the way you would size any model: the weights (parameters × bits ÷ 8), the KV cache for 8,192 tokens of context, and roughly 1.5 GB for the runtime itself. Real use varies with settings.

Each token of context adds about 0.04 MB of KV cache at 16-bit, so 32,768 tokens would add about 1.3 GB. 25 of its 30 layers use sliding-window attention, so the real cache is smaller than this full-attention estimate.

How the verdicts are worked out: on ANVE One, about three quarters of 64 GB of unified memory is available to the model; on ANVE Office and ANVE Office Pro, the model has to fit in the 24 GB on the graphics card, with 64 GB or 128 GB of system memory for everything else. This is arithmetic against the launch specification, not a test. ANVE’s tested default is Qwen 3.8 27B; this model is not yet validated on ANVE.

This is the class ANVE runs by default (27B-class): strong enough for meeting minutes, summaries, drafting and question answering over an office’s own documents, and small enough for one box.

• Qwen3.8-27B (27.8B parameters, about 17 GB at 4-bit before context; commercial use allowed)
• Qwen3.6-27B (27.8B parameters, about 17 GB at 4-bit before context; commercial use allowed)
• Qwen3-VL-30B-A3B-Instruct (31.1B parameters, about 19 GB at 4-bit before context; commercial use allowed)

No sealed ANVE unit exists; every image of the appliance on this site is a labelled render. The software runs today on a development machine. No customers yet, no certifications, no delivery date, and the final price is not announced.

Reservations are ₹999 in India or US$12 elsewhere, refundable in full until you confirm an order in writing (terms). Offices that cannot use cloud AI can apply to be one of ten design partners.