What is IPFS?
The InterPlanetary File System (IPFS) is a peer-to-peer protocol that lets nodes store and transfer files between one another. It is a communication protocol, not a network — software like the IPFS Desktop client or the IPFS CLI daemon implements the protocol and creates a network of peers that store and share files.
IPFS vs HTTP
HTTP, the protocol underpinning the modern web, follows a client-server model: your browser asks one specific server for a resource, and if that server is unavailable, the request fails.
IPFS uses a peer-to-peer model: any node holding a copy of a file can serve it to any other node. Pulling content by CID lets you fetch from whichever peer responds first, which provides resilience against single-server outages.
To bridge the two protocols, IPFS HTTP gateways accept normal HTTP requests for IPFS CIDs and serve the content back over HTTP — letting browsers and HTTP-only tools work with IPFS content.
Content addressing
Data on IPFS is identified by content address — a hash of the file's bytes — rather than by location. This means:
- The same file uploaded twice produces the same CID.
- Any change to the bytes produces a different CID.
- A CID is a self-verifying handle: anyone fetching content by CID can hash what they received and confirm it matches.
The CID is stored as part of every Filebase IPFS object's metadata.
Native IPFS URLs
Applications that natively support IPFS can address content using:
ipfs://{CID}/{optional path}
For tools that only speak HTTP, use a gateway:
https://ipfs.filebase.io/ipfs/{CID}
Pinning
By default, IPFS nodes garbage-collect cached content over time. Pinning marks content as non-collectible — guaranteeing it stays available on the pinning node.
Filebase's IPFS service automatically pins every object you upload to an IPFS bucket. See pinning.