po-lang-engine 2.4.0

Po-Lang — Native LLM + ML + HTTP gateway programming language engine

Po-Lang Engine v2.4.0

A high-performance programming language with native AI/ML + LLM training engine.

pip install po-lang-engine
pop run script.po

v2.4.0 — Native LLM Training Engine

Train language models directly in Po-Lang scripts — no Python glue code needed.

Size	Model	Speed	Use case
"nano"	N-gram (n=4)	Instant	Quick text generation
"micro"	CharRNN 64-dim	Pure Python	Any device (Termux)
"small"	CharRNN 256-dim	numpy 30x	Fast training

use ai
use net
import sys

keep llm = ai.LLM("micro")

if sys.exists("models/lang.po_model") {
    llm["load"]("models/lang.po_model")
    sys.stdout("[LLM] Model loaded!\n")
} else {
    keep corpus = "Po-Lang is a fast language. Po-Lang runs anywhere."
    llm["train"](corpus, 50)
    llm["save"]("models/lang.po_model")
}

keep text = llm["generate"]("Po-Lang", 100, 0.8)
show text

HTTP AI Gateway

keep server = net.listen(8080)
server["route"]("GET", "/generate", fn(req) {
    keep prompt = req["query"]("prompt")
    keep text   = llm["generate"](prompt, 150, 0.8)
    return "{ 'generated': '" + text + "' }"
})
server["start"]()