🆘 Got errors? Talk to me live — straight from this machine
You can't copy-paste between your machines, so here's a direct line to me. Open the bridge, paste whatever error or command output you're seeing, and I'll reply there with exactly what to run next. It auto-refreshes — replies land in ~30s while I'm watching.
It only builds a form if its base model is pulled, and prints exactly what it made.
d · Chat box says "no connection"? Let Ollama accept the page once, then quit & reopen Ollama:
setx OLLAMA_ORIGINS "*"
e ·killak-ssgss (32B) painfully slow or stalling? That's the big one filling your 4090 — just use killak-ssj (14B) as your daily. Still anything weird → 🆘 bridge above.
🚫 Kill YouTube ads on this PC
Heads up — your hosts-file trick was never gonna work, and that's not on you: YouTube serves its ads from the same domains as the actual videos, so blocking by domain either does nothing or kills YouTube completely. The only thing that reliably works is an ad-blocking browser. Pick one:
Easiest (recommended) — Brave. A browser that blocks YouTube ads out of the box, zero setup. Install it, then just watch YouTube in Brave:
winget install Brave.Brave
Open Brave from the Start menu → go to youtube.com → sign into your Google account (so your subs + algorithm carry over) → ads gone, including those 10-min ones. Done.
Most bulletproof — Firefox + uBlock Origin. If Brave ever slips, this combo is the gold standard and updates fastest against YouTube's tricks:
winget install Mozilla.Firefox
Open Firefox → go to the uBlock Origin page → Add to Firefox → Add. Then watch YouTube in Firefox.
Why not just patch Chrome? Google's kneecapped ad blockers in Chrome (Manifest V3) — the Chrome version (uBO Lite) is much weaker on YouTube now. Brave or Firefox is the reliable route. Set whichever you pick as your default browser and you'll never think about it again.
🟢 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 killak-ssj
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. (Not upgraded yet? Use ollama run killa until you've run the 🧠 box below.)
🧠 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:
4 · Open the chat box → the dropdown now shows Killak / Killak (SSJ) / Killak (SSGSS), and the whole box recolours to match the form — grey → gold → blue ⚡. A/B them, tell me which wins. 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:
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 the Killak Super-Saiyan ladder — killak (base) + killak-ssj + killak-ssgss for whichever base models you've pulled.
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.
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
Fresh start? Close every PowerShell window, then do these in order, top to bottom. Two windows stay open at the end — the server + the public link. Runs on THIS 4090, so it uses your GPU + killak models.
🆕 Name + logging built in. After the password, each person types their name, and every message + reply is saved to chat-log.jsonl in the server folder (private, stays on this 4090) — your training-data firehose.
1 · One-time setup skip if already done
Installs Node + the tunnel tool. "Already installed" = grand. Then close & reopen PowerShell so it finds them.
Grabs the latest server and starts it. It will ask you for a password — type the one you'll give your mate. Leave this window open.
$Host.UI.RawUI.WindowTitle='(1) LLAMA SERVER - keep me open'; curl.exe -L https://llama-chat-8e6.pages.dev/llama-chat-server.zip -o "$env:USERPROFILE\Downloads\llama-chat-server.zip"; Expand-Archive "$env:USERPROFILE\Downloads\llama-chat-server.zip" -DestinationPath "$env:USERPROFILE\Downloads\llama-chat-server" -Force; cd "$env:USERPROFILE\Downloads\llama-chat-server"; $env:CHAT_PASSWORD=(Read-Host 'Type a password to give your mate'); node serve.mjs
You should see 🦙 Llama chat server is live (password ON). Quick check on this PC: 🔎 open localhost:8080 — should show the 🦙 password page.
3 · Window 2 — open the public link NEW PowerShell
Open a fresh PowerShell window and run this. It prints a https://…trycloudflare.com link. Leave it open too.
$Host.UI.RawUI.WindowTitle='(2) PUBLIC LINK - keep me open'; cloudflared tunnel --protocol http2 --url http://localhost:8080
Send your mate that link + the password you typed. Tell him to pick killa (or killak-ssj) in the model box — that's your GPU AI.
Keep BOTH windows open + Ollama running (it auto-starts with the PC). The link is fresh every run — always send the newest (an old one 404s). Link + password = a key; only give it to people you trust.
5 Remote into this 4090 from anywhere laptop / iPad
Tailscale tunnel is live between your machines. Do this once on the 4090 to allow Remote Desktop, then drive its screen from your laptop or iPad on your Tailscale — like sitting here.
⚠️ Your PIN does NOT work over Remote Desktop — it needs a real password. Two ways in: ① Take over YOUR desktop (all your open work) = connect as your own account + its password. If you sign into this PC with an email, that's your Microsoft-account password — reset it at account.microsoft.com if you only ever use the PIN and forgot it. (Do step b only; skip a.) ② A clean separate desktop (your files + apps, fresh session) = make the rdp account in step a and connect as that.
a · Make a dedicated remote login admin PowerShell
Right-click Start → Terminal (Admin). Change CHANGE_ME_pass to a password you'll remember, then run:
net user rdp "CHANGE_ME_pass" /add; net localgroup Administrators rdp /add; net localgroup "Remote Desktop Users" rdp /add
Remote Desktop app → this address, user rdp, the password you set. (iPad: install the RD Client + Tailscale apps, same account.)
100.103.53.92
Keep the 4090 on + plugged in. Tailscale's already on your account on both machines.
6 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.
Open README.md in there and give it a read — it's the honest version of all this.
Step 2 — Set up the trainer oncefiddly
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.