Unlimited Context
Local-first virtual memory engine that lets small local LLMs reach a billion-token disk pool in slices.
Tested on Linux
Use it
Linux x86_64
You need first
- Runtimes:
Python 3.10+
Install via uv tool (candidate 1)
uv tool install aether-context
Check that it works
aether-context --version 2>&1The output should contain “aether-context 0.5.0”.
First thing to try
aether-context setup 2>&1Run aether-context setup first to check that the engine works.
What it does
- Keep a billion-token memory pool on local diskTested
How to use it
- Run
aether-context setupto check that the engine can store and retrieve slices. - Run
aether-context statusto see the pool size and how many tokens it can reach.
- Run
- Recall a connected thread through an MPO chainNot shown working
- Use any local model through OllamaNot tested
How far pickfit got
Linux x86_64
- Installs
- Starts
- 1 of 3 features work
- Works offline
Good for
- remember facts across runs in a local pool
- recall stored facts before generation
- run long coherent coding agent tasks
- chat interactively with full memory reach
- check local model, disk and RAM before a run
How pickfit tested it
Each run starts from a clean machine and follows the steps above. How pickfit verifies
Linux x86_64
Its main function works
- Replayed from scratch
- Works offline
- Checked on 30 Oct 2026
- Version 0.5.0
Differences the test machine hit
- README: pip install aether-context
In the test: uv tool install aether-context (isolated, avoids PEP 668 / read-only system)