Metadata-Version: 2.4
Name: pankhllm-server
Version: 0.1.1
Summary: The LLM gateway that learns to skip the LLM: trains its own tiny decision model on your agent's traffic and answers repeat decisions in 0.2 ms with no LLM call. OpenAI-compatible, one Rust binary. The server binary for pip users: pankhllm serve
Keywords: llm,gateway,router,openai,agents
Home-Page: https://singhpratech.github.io/pankhllm/
License: MIT
Requires-Python: >=3.8
Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
Project-URL: Homepage, https://singhpratech.github.io/pankhllm/
Project-URL: Repository, https://github.com/singhpratech/pankhllm

# pankhllm-server

The [pankhllm](https://github.com/singhpratech/pankhllm) router binary, installed with pip.
pankhllm is the LLM gateway that learns to skip the LLM: it trains its own tiny decision model
on your agent's traffic and answers repeat decisions in about 0.2 ms on a CPU with no LLM
call. Everything unfamiliar still goes to the LLM you already use. OpenAI-compatible.

```sh
pip install pankhllm-server
pankhllm serve --port 4000      # then point any OpenAI client at http://localhost:4000/v1
```

The pure-Python client and executor helper are in the `pankhllm` package:
`pip install pankhllm`. Live demo and docs: https://singhpratech.github.io/pankhllm/

