Open weights
Gemma 4
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
- Vision
- Tool calling
AI model category
Open weights AI models
For teams that need to run models on their own infrastructure, for control, privacy or cost.
- 13 models
Open weights models can be downloaded and run on your own servers or a cloud provider of your choice. That gives full control over data and versions, and can lower cost at high volume. It also means you take on hosting, scaling and updates yourself.
Licenses differ. Some are fully permissive, while others add conditions for large companies or certain uses. Always read the license before building on a model.
Open weights
All open weights ai models
Image and video
Stable Diffusion 3.5
Stability AI
Stability AI's open-weight image models, in Large, Medium and Flash versions, which can be self-hosted or used through the Stability API.
- Stability AI
- Open weights
- Vision
Image and video
FLUX.2
Black Forest Labs
Black Forest Labs' FLUX.2 image models: API-only Pro, Flex and Max versions, plus open-weight Dev and Klein versions under different licenses.
- Black Forest Labs
- Open weights
- Vision
Open weights
Llama 4 Scout
Meta
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.
- Meta
- Open weights
- Vision
- Tool calling
Open weights
Llama 4 Maverick
Meta
The larger Llama 4 open-weight model, with image input and a 1 million token context window, for self-hosted assistants and analysis.
- Meta
- Open weights
- Vision
- Tool calling
Open weights
Llama 3.3 70B
Meta
An older, text-only open-weight Llama model with tool use and a 128K token window, still widely hosted and well understood.
- Meta
- Open weights
- Tool calling
General purpose
Mistral Large 3
Mistral AI
Mistral's open-weight general-purpose flagship under Apache 2.0, with image input, tool calling, structured outputs and a 256K token window.
- Mistral AI
- Open weights
- Vision
- Tool calling
Open weights
Mistral Small 4
Mistral AI
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.
- Mistral AI
- Open weights
- Tool calling
Coding
Mistral Medium 3.5
Mistral AI
A newer open-weight Mistral model for agent and coding work, with image input, tool calling, structured outputs and a 256K window.
- Mistral AI
- Open weights
- Vision
- Tool calling
General purpose
DeepSeek V4.1 Flash
DeepSeek
DeepSeek's default, best-value model: open weights under MIT, image input, tool calling, JSON output, optional thinking and a 1 million token window.
- DeepSeek
- Open weights
- Vision
- Tool calling
Reasoning
DeepSeek V4 Pro
DeepSeek
DeepSeek's model for demanding reasoning and agent work, with thinking effort levels, tool calling, JSON output and a 1 million token window.
- DeepSeek
- Open weights
- Tool calling
Coding
Qwen3-Coder
Alibaba Cloud
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.
- Alibaba Cloud
- Open weights
- Tool calling
General purpose
Cohere Command A+
Cohere
Cohere's enterprise flagship for multimodal, multilingual agent tasks, with 48 languages, tool calling, JSON output and open weights under Apache 2.0.
- Cohere
- Open weights
- Vision
- Tool calling
How to choose
How to choose in this category
Compare the best open model you can host against a hosted API model on the same test set. Include the real cost of GPUs, engineering time and monitoring. Self-hosting usually makes sense for strict data rules, very high volumes or offline use, and less often for small teams.
Other categories
Browse other categories
Keep exploring
Related on ExecMedia
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.
Solutions
View all solutionsIndustries
View all industriesGuides 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.
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.
- 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.
- OneDrive Microsoft's OneDrive MCP server, part of its Work IQ preview, lets AI find, read, create, move, share and delete files in OneDrive.
Glossary terms
View all glossary terms- 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.
- 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.
- Inference Inference is the step where a trained AI model is used to produce an output, such as an answer, a label or a prediction, from new input.
- LLM integration LLM integration is connecting a large language model to your software and data, so AI features work inside your own products and processes.
- Prompt engineering Prompt engineering is the practice of designing, testing and improving prompts so AI models produce reliable results for a specific task.
- Token A token is a small piece of text, often part of a word, that AI models read and write, and that providers use to measure limits and pricing.