GitHub - philipl/inferencefs: The new data-free filesystem! · GitHub
/" data-turbo-transient="true" />
Skip to content
Search or jump to...
Search code, repositories, users, issues, pull requests...
-->
Search
Clear
Search syntax tips
Provide feedback
--><br>We read every piece of feedback, and take your input very seriously.
Include my email address so I can be contacted
Cancel
Submit feedback
Saved searches
Use saved searches to filter your results more quickly
-->
Name
Query
To see all available qualifiers, see our documentation.
Cancel
Create saved search
Sign in
/;ref_cta:Sign up;ref_loc:header logged out"}"<br>Sign up
Appearance settings
Resetting focus
You signed in with another tab or window. Reload to refresh your session.<br>You signed out in another tab or window. Reload to refresh your session.<br>You switched accounts on another tab or window. Reload to refresh your session.
Dismiss alert
{{ message }}
philipl
inferencefs
Public
Notifications<br>You must be signed in to change notification settings
Fork
Star<br>14
main
BranchesTags
Go to file
CodeOpen more actions menu
Folders and files<br>NameNameLast commit message<br>Last commit date<br>Latest commit
History<br>42 Commits<br>42 Commits
.github/workflows
.github/workflows
src/inferencefs
src/inferencefs
tests
tests
.gitignore
.gitignore
.python-version
.python-version
CLAUDE.md
CLAUDE.md
LICENSE
LICENSE
README.md
README.md
appendix.md
appendix.md
plan.txt
plan.txt
pyproject.toml
pyproject.toml
View all files
Repository files navigation
InferenceFS: Never worry about data again! (Again!)
The long-awaited successor to πfs.
Over a decade ago, πfs proved that you didn't need to store your data —<br>it was all there in the digits of pi, waiting to be looked up. The only<br>thing you needed to keep was metadata: the offsets into pi where your<br>bytes happened to live.
πfs was ahead of its time. The metadata was, regrettably, larger than<br>the data it replaced. Moore's Law was supposed to fix this but got<br>distracted. The project needed a breakthrough — a way to store even<br>less and still get your data back.
That breakthrough is InferenceFS .
Where πfs looked up your data in a transcendental number, InferenceFS<br>looks it up in something far more powerful: the latent space of a large<br>language model trained on the entire internet. Where πfs stored byte<br>offsets as metadata, InferenceFS stores only the filename. That's it.<br>The filename is the metadata. The contents are inferred.
πfs required metadata proportional to your data. InferenceFS requires<br>metadata proportional to your filenames. This is the compression<br>breakthrough the industry has been waiting for.
And the timing couldn't be better. In today's economy, where a stick of<br>DDR5 costs more than a used car and NVMe drives are priced per byte like<br>saffron, InferenceFS offers a radical alternative: store terabytes of<br>data using only kilobytes of filenames . Copy your entire home directory<br>into InferenceFS and watch your disk usage plummet to nearly zero. The<br>contents are still there — they're just held in the parametric memory of<br>a model with billions of weights that someone else is paying to host.
What's changed since πfs?
πfs<br>InferenceFS
Data stored in<br>The digits of π<br>The latent space of an LLM
Metadata per file<br>Byte offsets (larger than the file)<br>Just the filename
Compression ratio<br>Negative<br>Technically infinite
Data retrieval<br>Bailey-Borwein-Plouffe formula<br>API call ($0.01)
Theoretical basis<br>π is conjectured to be normal<br>LLMs are conjectured to be useful
Speed<br>Five minutes for a 400-line file<br>Five seconds for a 400-line file
Cloud ready<br>No<br>Inherently — someone else's GPU is doing all the work
Features
Unlimited effective storage capacity — bounded only by your imagination<br>(and your API bill)
Aggressive deduplication — all files with the same name share the same<br>contents, guaranteed
Transparent operation — mounts as a standard filesystem, works with<br>cat, ls, cp, and every other tool
Multiple LLM backends — Claude API, Claude Code CLI, and Google Gemini
Built-in content cache with configurable LRU eviction
Full binary file support — generates valid PNG, JPEG, GIF, PDF, WAV,<br>MP4, ELF, and PE files with correct magic bytes and headers
Written file sizes are persisted across mount/unmount cycles
Unlike πfs, actually returns plausible content in finite time
Quick Start
We recommend the Gemini backend for the best experience — it's the<br>fastest (sub-second response times), has the widest binary format<br>support, and offers a generous free tier. Get an API key at<br>ai.google.dev.
# Install from PyPI<br>pip install inferencefs<br># or<br>uv pip install inferencefs
# Or install from source<br>git clone https://github.com/philipl/inferencefs.git<br>cd inferencefs<br>uv sync
# Mount with Gemini backend (recommended)<br>export API_KEY=your-key-here<br>mkdir -p /tmp/source /tmp/mount<br>uv run inferencefs --backend gemini /tmp/source /tmp/mount
# Create some files and read them<br>touch...