44. Octo — model shell

Part: VII · Apps
Package: github.com/openfluke/welvet/apps/octo
Status: ok — ✅ runs

When

You need the apps/octo foundation package.

Where

import "github.com/openfluke/welvet/apps/octo"

cd 44-octo && source ../env.sh && go run .

Why

A model is only useful with a shell around it: pull weights from Hugging Face, convert them to a Welvet .entity, then chat, serve, or benchmark. Octo is that shell, kept in its own module so the engine never depends on an app.

What

Subcommands cover the whole loop: hub download/ensure repos, convert pack to a single .entity, interactive run/serve/chat, plus image and speech menus. Its bench harness sweeps every quant format across a CPU Plan 9 SIMD fused profile and a WebGPU fused profile.

Sample output (captured)

Octo is its own module — the engine never imports it.
  cd apps/octo && go run .
  download -> convert -> .entity -> chat / serve / bench

Live capture from go run ./cmd/runall on local ../../welvet (exit 0).

Source

Copied from the Welvet feature book examples (openfluke.github.io/welvet/examples/44-octo).

package main

import "fmt"

func main() {
	fmt.Println("Octo is its own module — the engine never imports it.")
	fmt.Println("  cd apps/octo && go run .")
	fmt.Println("  download -> convert -> .entity -> chat / serve / bench")
}