🦙 Killa's Local AI — Control Panel

🟢 Just booted your PC? Start chatting in 10 seconds

The AI engine (Ollama) installs itself as a Windows service — so it's already running the second your PC turns on. You never "start the engine". You just open a way to talk to it. Fastest way, straight into PowerShell:

ollama run killa

Type away · /bye to leave. First message after a boot takes a few seconds (it's loading your brain into the 4090). Want the pretty chat box or your phone? → sections 2 & 3.

🧠 Make killa way smarter — tonight's upgrade once

Two fixes in one: killa stops parroting your slang back at you, and it stops confidently making things up. Your 4090's been running a small model in first gear — this drops it into top gear (bigger model = knows more, bluffs less).

Run these in order — one-time. Big downloads, but only the once.

1 · Pull the two bigger brains — 14B (~9GB) + 32B (~20GB). Stick the kettle on:

ollama pull qwen2.5:14b; ollama pull qwen2.5:32b

2 · Grab tonight's updated brain files:

curl.exe -L https://llama-chat-8e6.pages.dev/killa-brain.zip -o "$env:USERPROFILE\Downloads\killa-brain.zip"; Expand-Archive "$env:USERPROFILE\Downloads\killa-brain.zip" -DestinationPath "$env:USERPROFILE\Downloads\killa-brain" -Force

3 · Build both models from your brain — killa (14B, daily) + killa-32b (32B, max):

powershell -ExecutionPolicy Bypass -File "$env:USERPROFILE\Downloads\killa-brain\build.ps1"

4 · Open the chat box → pick killa (fast) or killa-32b (smartest) from the model dropdown. A/B them, tell me which wins. The old llama3.1:8b can stay or be binned later — your call. Still £0 — all on your GPU.

😴 Too tired to read? Tonight's update in 20 seconds

What changed: when your mates open your shared link, after the password they now type their name — and every question + answer gets saved on your 4090 (private, never leaves). That's you banking real training examples while they chat.

What your mates do: nothing new — just type a name once. They're never asked the password or name again after that.

The ONE thing YOU do to switch it on: the server you've got running is the old one. Grab the upgraded one and restart it — that's it. Full steps in Section 4, or the quick grab:

curl.exe -L https://llama-chat-8e6.pages.dev/llama-chat-server.zip -o "$env:USERPROFILE\Downloads\llama-chat-server.zip"; Expand-Archive "$env:USERPROFILE\Downloads\llama-chat-server.zip" -DestinationPath "$env:USERPROFILE\Downloads\llama-chat-server" -Force

Safe to run — it upgrades the code but keeps your password and your saved chat log. Then start it + share the link from Section 4.

💷 Does any of this cost money? No — £0. The dashboard's on free Cloudflare Pages, the share-link is a free tunnel, and the AI runs on your own 4090. No Cloudflare Workers, no bills, no limits you'll hit. Only "cost" is the electric running your GPU.

1 First time on this PC once

Setup steps. If something's already done, skip it — running twice does no harm.

Install Ollama once

The engine that runs the AI. (You've likely done this — skip if ollama --version works.)

winget install Ollama.Ollama

Then close and reopen PowerShell.

Get the model once

Downloads the brain-weights. qwen2.5:14b (~9GB) is the daily driver; the 32B (~20GB) is optional for max quality.

ollama pull qwen2.5:14b; ollama pull qwen2.5:32b

In a hurry / tight on disk? Just the first one is enough to get going.

Give it YOUR brain → the killa model once

Bakes in your identity + tone so it knows it's local and knows you. Builds killa (14B) and, if you pulled it, killa-32b (32B).

curl.exe -L https://llama-chat-8e6.pages.dev/killa-brain.zip -o "$env:USERPROFILE\Downloads\killa-brain.zip"; Expand-Archive "$env:USERPROFILE\Downloads\killa-brain.zip" -DestinationPath "$env:USERPROFILE\Downloads\killa-brain" -Force; powershell -ExecutionPolicy Bypass -File "$env:USERPROFILE\Downloads\killa-brain\build.ps1"

Test it: ollama run killa "who are you and where do you run?"

Install Node + tunnel tool once

Only needed for the wifi chat + mum's link (sections 3 & 4).

winget install OpenJS.NodeJS.LTS; winget install Cloudflare.cloudflared

Then close and reopen PowerShell.

Open THIS dashboard automatically every boot once

Drops a shortcut in your Startup folder so this page pops open in your browser every time you log in — no more hunting for it. Run once.

$u='https://llama-chat-8e6.pages.dev'; $s=[Environment]::GetFolderPath('Startup'); Set-Content "$s\Killa 4090 Dashboard.url" "[InternetShortcut]`r`nURL=$u" -Encoding ASCII; Start-Process $u

Changed your mind? Delete Killa 4090 Dashboard.url from shell:startup (paste that into the Run box / File Explorer bar).

2 Chat on this PC anytime

Quick terminal chat

Fastest. Talk to your brain-model right in PowerShell.

ollama run killa

Type away. /bye to exit.

Or the nice chat box (GUI)

Start the server (section 3 below), then open localhost:8080 — that page can actually talk to your AI. Pick killa in the model box.

⚠️ Don't chat on this web page — it's just your menu, it can't reach the AI (that's the 404 you saw). Chatting happens at localhost:8080 or in the terminal above.

💬 Open http://localhost:8080

3 Chat from any device on your wifi phone / tablet

Runs a little server here so your phone (or mum's, when she's round) can use it too.

Get the server once

curl.exe -L https://llama-chat-8e6.pages.dev/llama-chat-server.zip -o "$env:USERPROFILE\Downloads\llama-chat-server.zip"; Expand-Archive "$env:USERPROFILE\Downloads\llama-chat-server.zip" -DestinationPath "$env:USERPROFILE\Downloads\llama-chat-server" -Force

Start it each time

cd "$env:USERPROFILE\Downloads\llama-chat-server"; powershell -ExecutionPolicy Bypass -File .\start-here.ps1

It prints an address like http://192.168.1.50:8080 — open THAT on any device on the same wifi. Leave the window open while you use it.

4 Send a private link (mum / your mates) off your wifi

Two windows — one runs the server (with a password), one opens the public link. This way you can SEE it working (the old one-click hid its errors).

🆕 Now with name + logging. After they type the password they get asked their name — so you know who said what. And every message + reply is saved locally to chat-log.jsonl in the server folder (on this 4090, private, never leaves). That's your training-data firehose — the more your mates chat, the more real examples you bank. Peek at it any time with the command in ⚡ Handy commands.

First — grab the upgraded server do once after tonight's update

The names + logging live in the new server code. Run this to upgrade your copy (keeps your password + saved log).

curl.exe -L https://llama-chat-8e6.pages.dev/llama-chat-server.zip -o "$env:USERPROFILE\Downloads\llama-chat-server.zip"; Expand-Archive "$env:USERPROFILE\Downloads\llama-chat-server.zip" -DestinationPath "$env:USERPROFILE\Downloads\llama-chat-server" -Force

Window 1 — start the server with a password

Change mumpass123 to any password you like. Leave this window open.

cd "$env:USERPROFILE\Downloads\llama-chat-server"; $env:CHAT_PASSWORD="mumpass123"; node serve.mjs

You should see 🦙 Llama chat server is live (password ON).

Check it works HERE first

On this 4090, open localhost — you should get the 🦙 password page. If you don't, the server didn't start; tell me what Window 1 says.

🔎 Open http://localhost:8080

Window 2 — open the public link NEW PowerShell

Open a fresh PowerShell window and run this. It prints a https://…trycloudflare.com link.

cloudflared tunnel --url http://localhost:8080

Send mum that link + the password. Tell her to pick killa in the model box.

Keep BOTH windows open, and Ollama running. The link is fresh every time you run it — always send the newest one (an old link gives a 404). Link + password = a key; only give them to people you trust. Want a permanent link + per-person login? That needs a domain — just ask.

5 Training Lab — make it truly yours advanced

This is real training — baking your voice into the model's weights (a LoRA), not just notes it re-reads. Think muscle memory instead of a briefing. Do the steps in order.

Straight talk before you burn a night on this: fine-tuning locks in tone + identity — it does not make it know more facts or stop it making things up (that's model size, different job). And it's only as good as the data. The kit ships 50 seed examples — enough to nudge the voice + prove the pipeline, not transform it. We grow it toward ~300 (shadow-log + more) before it really bites. We keep the old killa until the new one earns the name.

Step 1 — Get the training kit once

The dataset, the training script, and the guide. Downloads to your Downloads folder.

curl.exe -L https://llama-chat-8e6.pages.dev/killa-train.zip -o "$env:USERPROFILE\Downloads\killa-train.zip"; Expand-Archive "$env:USERPROFILE\Downloads\killa-train.zip" -DestinationPath "$env:USERPROFILE\Downloads\killa-train" -Force; explorer "$env:USERPROFILE\Downloads\killa-train"

Open README.md in there and give it a read — it's the honest version of all this.

Step 2 — Set up the trainer once fiddly

Fine-tuning wants Linux tooling. Cleanest route on Windows = WSL2 — a real Ubuntu running inside Windows with your 4090 passed through. Think a Linux box in a window, sharing your GPU.

wsl --install

Reboot after this. Then open Ubuntu from the Start menu and run the next one inside it:

nvidia-smi && pip install unsloth

nvidia-smi should list your 4090 (proves the GPU's visible). pip install unsloth pulls the whole trainer + downloads the base model (~5GB) on first run. This is the bit that can fight you — if it errors, copy me the exact red line and I'll unstick it. Not pretending it's one-click.

Step 3 — Run the fine-tune ~30 min

Inside Ubuntu, go to the kit folder and train. Your 4090 does the work.

cd /mnt/c/Users/$USER/Downloads/killa-train && python train.py

It trains, then spits out killa-tuned/ — a ready-to-load model. Watch the loss tick down. Leave it cooking.

Step 4 — Load it & judge it head-to-head

Bake the result into Ollama as killa2, then ask both the same thing and see who's better.

cd /mnt/c/Users/$USER/Downloads/killa-train; ollama create killa2 -f killa-tuned/Modelfile; ollama run killa2 "who are you and where do you run?"

Then compare: ollama run killa "who are you and where do you run?". If killa2 wins, tell me and we promote it to killa. If not, we grow the data and go again. Proof, not faith.

⚡ Handy commands

Free the GPU before gaming

Dumps the model out of VRAM instantly.

ollama stop killa

See your models

ollama list

See what's loaded (GPU?)

ollama ps

Try the bigger model

Heavyweight Llama — slower, partly on RAM.

ollama pull llama3.3:70b

VS Code coding assistant

Then press F5 in VS Code.

curl.exe -L https://llama-chat-8e6.pages.dev/llama-code.zip -o "$env:USERPROFILE\Downloads\llama-code.zip"; Expand-Archive "$env:USERPROFILE\Downloads\llama-code.zip" -DestinationPath "$env:USERPROFILE\Downloads\llama-code" -Force; code "$env:USERPROFILE\Downloads\llama-code"

Teach the brain more

Edit brain.md in the killa-brain folder, then:

cd "$env:USERPROFILE\Downloads\killa-brain"; powershell -ExecutionPolicy Bypass -File .\build.ps1

See what your mates asked

Opens the chat log (who / prompt / reply) collected from the shared link.

notepad "$env:USERPROFILE\Downloads\llama-chat-server\chat-log.jsonl"