> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.recursiv.io/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.recursiv.io/_mcp/server.

# Recursiv vs OpenAI Assistants

## How Does Recursiv Compare to OpenAI Assistants?

OpenAI Assistants run on one provider. They give you an assistant with tools and a thread, but the surrounding backend is something you bring. Recursiv is model agnostic and a full backend with govern: any model, plus data, storage, auth, orchestration, deploy and the agent runtime.

The differentiator is portability and scope. With Recursiv you are not locked to a single model vendor and you get the backend and runtime around the agent, not only the assistant primitive.

## Quick Comparison

| Capability        | Recursiv                                     | OpenAI Assistants          |
| ----------------- | -------------------------------------------- | -------------------------- |
| Model support     | Any model, swap anytime                      | OpenAI models              |
| Agents and swarms | Agents and swarms                            | Single assistant           |
| Memory            | Brain and memory built in                    | Threads                    |
| Orchestration     | Dispatcher and jobs                          | Bring your own runtime     |
| Code execution    | Managed sandboxes                            | Code interpreter tool      |
| Data and storage  | Postgres and object storage                  | Bring your own             |
| Auth              | Built in                                     | Bring your own             |
| Deploy            | Deployments built in                         | Bring your own host        |
| Integrations      | Hundreds built in                            | Function calling and tools |
| Govern            | Scoped identity, approvals, audit, self-eval | Build it yourself          |

## When OpenAI Assistants Are the Better Choice

* You are committed to OpenAI models and do not need model portability
* You want a single assistant primitive and already have your backend
* Your use case fits threads and tool calls without orchestration, memory or govern
* You want to stay entirely within the OpenAI ecosystem

## Get Started

If you want to stay model agnostic and get a full backend with govern around your agents, start with the [quickstart](/guides/start-here/quickstart).