AI model by Google
Gemma 4 for business
Google's open-weight model family under Apache 2.0, in sizes from phone-friendly to 31B, with image input and function calling.
- Open weights
- General purpose
- Vision and multimodal
- Open weights
- Reviewed on September 24, 2026
Capability tiersRelative, not benchmarks
Key facts about Gemma 4
- Model ID at review
- Gemma 4 (E2B, E4B, 12B, 26B A4B, 31B)
- Provider
- Main category
- Open weights
- Open weights
- Yes, Apache 2.0
- Last reviewed
- September 24, 2026
Fit
Where it fits and where it does not
Good at
-
Running on your own servers or devices.
-
Keeping data inside your own environment.
-
Image input on all sizes, audio on the smaller ones.
-
Function calling for simple agents.
-
Many languages.
Not the right choice for
-
The hardest reasoning tasks, where large hosted models lead.
-
Teams without the skills or budget to run models.
-
Very long contexts beyond 256K tokens.
-
Cases where a hosted API would be simpler and cheaper.
Use cases
Business use cases we would use it for
Private assistants
On-device features
Private classification
Our notes
When we would choose it
Gemma 4 is Google's current open-weight model family, released under the Apache 2.0 license. It comes in several sizes, E2B, E4B, 12B, 26B A4B and 31B, with context windows of 128K tokens on the smallest two and 256K on the larger ones. All sizes accept text and images, the smaller ones also accept audio, and Google lists native function calling and support for more than 140 languages.
Open weights mean you can download the model and run it where you like: your own servers, a cloud of your choice or even a phone. That gives full control over data and versions. The trade-off is that you run the infrastructure, handle scaling and apply updates yourself.
We would consider Gemma when data rules require that nothing leaves your environment, when an app needs to work offline, or when volumes are high enough that owning the hardware is cheaper than paying per request. For many small teams, a hosted model is still simpler.
The price level shown here reflects free weights, not the cost of hardware and people. Before deciding, compare the total cost and quality with a hosted option such as Gemini Flash-Lite. See open weights models.
A good first test is to run the 12B or 26B size on a single rented GPU against a sample of your real tasks, and compare the answers with a hosted model. That gives you quality and cost figures for your own work before you invest in hardware.
Before you commit
Things to check before you commit
-
Hardware
Estimate GPU or CPU needs for your chosen size and traffic.
-
Operations
Someone must run, update and monitor the model.
-
Quality gap
Compare with a hosted model on the same test set.
-
License
Gemma 4 uses Apache 2.0. Confirm it fits your use.
Alternatives
Models to compare it with
Llama 4 Scout
An open-weight Llama 4 model with image input and a very long context window, for teams that want to host a capable model themselves.
Mistral Small 4
Mistral's efficient open-weight model under Apache 2.0, one hybrid model for instructions, reasoning and coding, with tool calling and a 256K window.
Qwen3-Coder
Qwen's open-weight coding models under Apache 2.0, from the efficient Qwen3-Coder-Next to the large 480B model, with tool use and long context.
Keep exploring
Solutions, services and guides
Related services
View all related services- LLM integration Add a large language model to software you already have, with the guardrails, costs and logging handled.
- AI model evaluation Test candidate AI models on your real data and tasks, then pick the one that balances quality, speed and cost.
- AI Product Development AI agents, knowledge assistants, copilots and document automation built into the way your team already works.
- AI copilots AI helpers built into your product or internal tools that draft, summarize and suggest while people decide.
- Document automation Read invoices, forms, contracts and IDs, pull out the right fields and route them for review.
- Computer vision Software that reads photos and scans: damage checks, stock counts, document capture and quality control.
Solutions
View all solutions- Knowledge assistant (RAG) Ask a question in plain words and get an answer from your own documents, with links to the sources.
- AI document processing Read forms, applications, IDs and statements, extract the fields you need and route each document for the right review.
- AI-assisted content workflows Briefs, first drafts, edits and repurposing in a workflow where people set the angle and approve every word.
- AI translation workflows Translate websites, products and support content quickly with AI, a shared glossary and native-speaker review where it matters.
- Feedback analysis Read every review, survey and ticket, group them by theme and sentiment, and show what customers keep asking for.
- Screening assistant Summarize each application against your criteria, flag strong matches and missing information, and leave every decision to a person.
Industries
View all industries- Legal Contract review assistants, knowledge search, intake and document automation for law firms and in-house legal teams.
- Education and elearning Course platforms, student portals, content workflows and AI tutoring assistants for schools and training companies.
- Manufacturing Quality inspection, production dashboards, quoting tools, knowledge assistants and legacy system modernization for manufacturers.
Case studies
View all case studiesGuides and articles
View all guides and articles- How to choose an AI model for your business A step-by-step way to pick an AI model by testing candidates on your own tasks, data rules and budget.
- RAG vs fine-tuning Two ways to make AI work with your knowledge, compared by cost, accuracy and upkeep.
- What is RAG and when do you need it? How retrieval-augmented generation lets AI answer from your documents, when it fits and how to build it well.
- How to keep customer data safe when using AI Practical steps to protect customer data when you use AI services, from data terms to access and logging.
MCP servers
View all mcp servers- Files and storage Servers that connect AI to cloud drives and object storage, so it can find, read and organize documents.
- AWS S3 Amazon S3 is handled through the managed AWS MCP Server, which lets AI run scripted AWS actions, create presigned links and search AWS docs.
- BigQuery Google's remote BigQuery MCP server lets AI list datasets and tables and run SQL, with a read-only query tool and IAM controls.
- Box Box's official MCP server lets AI search and read files, ask questions across documents with Box AI, extract data and upload files.
- Dropbox Dropbox's own MCP server lets AI list, search and read files, convert them to markdown and create, move, share and delete files. In beta.
- Google Drive Google's own Drive MCP server lets AI search, read and create files in Google Drive, with your Google Workspace sign-in.
Glossary terms
View all glossary terms- Large language model A large language model, or LLM, is an AI model trained on vast amounts of text that can understand and generate language, and often images and code.
- Machine learning Machine learning is a way of building software that learns patterns from data to make predictions or decisions, instead of following only hand-written rules.
- Open-weights model An open-weights model is an AI model whose trained parameters are published, so anyone can download and run it on their own hardware under its license.
- Artificial intelligence Artificial intelligence is the broad field of building software that performs tasks that normally need human judgment, such as understanding language or images.
- Context window A context window is the maximum amount of text, measured in tokens, that an AI model can consider at once, including the question, documents and its answer.
- Fine-tuning Fine-tuning is further training an existing AI model on your own examples so it learns a specific style, format or task.
FAQ
Questions people ask us
Have a question that is not here? Ask us directly.
Gemma 4 is released under Apache 2.0, a permissive license. Read the license and terms for your use case.
E2B, E4B, 12B, 26B A4B and 31B, according to the Gemma documentation.
Yes, all sizes accept images. Audio input is available on the smaller sizes.