Ultra-low latency LLM routing based on Kolmogorov Complexity and Compression.
Project description
kolmo 🧠
Ultra-low latency LLM routing based on Kolmogorov Complexity and Statistical Compression.
kolmo is a zero-VRAM, microsecond-latency router for Large Language Models. Instead of using neural networks (embeddings) to classify user intent, it uses Information Theory and Normalized Compression Distance (NCD) to route prompts to the most efficient model.
🚀 The Vision
- Zero VRAM: Runs entirely on CPU, freeing up your GPU for what matters.
- Explainable: No black-box embeddings. Understand exactly why a prompt was routed based on compression ratios.
- Language Agnostic: Works at the byte level. Supports any human language or programming language natively.
- Blazing Fast: $O(n)$ complexity based on text length.
🛠 Installation
pip install kolmo
🏗 Status
kolmo is currently in Pre-Alpha. The current release (v0.0.1) is a placeholder for the upcoming information-theoretic routing engine.
Created by Carlos Bustamante & KODEX.
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file kolmo-0.0.1.tar.gz.
File metadata
- Download URL: kolmo-0.0.1.tar.gz
- Upload date:
- Size: 3.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.9.9 {"installer":{"name":"uv","version":"0.9.9"},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"22.04","id":"jammy","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
375abe3c59e3d5f123bd322d35e41eb998b525d4da1db893725ce1d2713828e0
|
|
| MD5 |
bcda77a6a797b3dfa416338b84c84860
|
|
| BLAKE2b-256 |
51792413305cf8f33a40ae842910146f32c9d142a0a3ea86570f4f67fbc21b1f
|
File details
Details for the file kolmo-0.0.1-py3-none-any.whl.
File metadata
- Download URL: kolmo-0.0.1-py3-none-any.whl
- Upload date:
- Size: 2.2 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.9.9 {"installer":{"name":"uv","version":"0.9.9"},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"22.04","id":"jammy","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
4f803931fc04467f8390c3a040662c7cf9dc025ead6b80cefaea08b81f6cc745
|
|
| MD5 |
d1825be24fe00ffd2ab3930b9da171ad
|
|
| BLAKE2b-256 |
b770a7c55973c89d9f513f09eb8471a96ec758896dd871736eb9b54e405d6d1e
|
