The narrator demonstrates a tower defense game and a spreadsheet app, both built with a free model collection he calls bark combo, made with a tool called Omni route that reaches over 50 providers, many needing no credit card.
OmniRoute is a GitHub repo with over 50,000 stars that you self-host, and its intelligent system picks providers with free quota to serve models into an IDE, giving roughly 1.5 billion tokens a month, or just over 2 billion in the first month with signup bonuses.
The two warnings are that these free API calls are not private, so only send information you are happy to share publicly, and some providers may not be happy about routing for free usage, so proceed at your own risk.
Installation is npm install g omni route from omni route.online followed by the OmniRoute command, which prints an API key and opens the dashboard; that key and base URL go into DeepSeek Harness as a custom provider named omni-route with a model set to auto.
After connecting OpenRouter and NVIDIA NIM free tiers, he builds a round robin combo named Bart dash combo from only the endpoints that test green — OpenRouter auto plus GPT OSS 120B and 20B and Nemotron 3 Super and Ultra — then has DeepSeek build a Pong replica in about 30 seconds, while warning that quality varies so he would not use it for anything production.
Brief overview
OmniRoute pools free provider quotas into one combo so DeepSeek Harness codes without paid APIs.
Pool many free providers into one combo endpointFive to fifteen connected providers plug into a single API call that DeepSeek Harness consumes.
Free usage costs privacy, not moneyOpenRouter's free models require the data-collection privacy toggles turned on, so whatever you send is shared.
Only keep endpoints that test greenBroken or throttled models break a coding session, so cherry pick the working ones and re-check periodically.
Questions this recording answers
10 questions, each answered where it is said
What is OmniRoute and how does it provide free AI model usage?
OmniRoute is a self-hosted GitHub repo with over 50,000 stars that connects to over 50 providers, many requiring no credit card. It intelligently determines which providers have free quota, picks those models, and serves them into your IDE, such as the DeepSeek harness, through one API.
How many free tokens can you get each month through OmniRoute?
Through the connected providers you can get approximately 1.5 billion tokens of usage continuously each month. Some providers give an additional sign-up bonus, so if you take advantage of those, your first month can reach just over 2 billion tokens of usage.
What are the risks of using free API calls through OmniRoute?
The free API calls are not private, so anything you send should be something you are happy to share publicly. Also, some providers might not be happy that you are routing and optimizing for free usage, so there are warnings and you do this at your own risk.
How do you install and start OmniRoute?
Go to omniroute.online, copy the command npm install g omniroute and paste it in a terminal; installation takes about five minutes. Then run the omniroute command to spin up the dashboard and server. The presenter suggests first asking ChatGPT or Claude to audit the code for safety.
How do you connect OmniRoute to the DeepSeek harness?
Copy the API key from the terminal, open DeepSeek settings, go to models and add a custom provider named omni-route. Paste the base URL, keep the OpenAI completions protocol, use a placeholder API key like 1234, and add a model set to auto for immediate free usage.
How do you set up OpenRouter as a free provider?
Sign up on OpenRouter, create a new API key with a useful name and copy it. In the privacy settings, turn on the options that are disabled by default, since free models collect your information. Then paste the key in OmniRoute, check it works, and import only free models.
Why add NVIDIA NIM as a provider?
NVIDIA has Nemotron models that are fast and very smart. You sign up for a free account, generate an API key in settings, paste it into OmniRoute and import free models. The GPT OSS 120 model tested successfully, and OpenRouter plus NVIDIA alone built the demo apps.
What is a combo and how do you build a reliable one?
A combo plugs all your providers into one API call for DeepSeek. The presenter named his bart-combo using the free stack template, kept OpenRouter's auto endpoint, and chose only working NVIDIA models: GPT-OSS 120B, 20B, Nemotron 3 Super and Nemotron 3 Ultra, using the default round robin strategy.
Why remove non-working models from the combo?
If you plug in all providers and the coding agent hits a model that does not work, that breaks the entire coding session. So you test each endpoint, cherry pick out the failing ones and leave only working models, though you should still check every so often.
Is this free setup good enough for production use?
No. With 10 or 20 different models plugged in, quality varies: one call might get a big smart model and another a small, less smart one. The presenter would not use it for anything production, only for fun, though a Pong game was built in about 30 seconds.
Key Quote
“In this video, I'm going to show you how to use the DeepSeek Harness completely for free,”
— Speaker
Key Quote
“This is probably the best tower defense game that I've built.”
— Speaker
Key Quote
“OmniRoute is a tool that will be self-hosting on our computer, and we'll be plugging in a bunch of different providers directly into OmniRoute.”