OpenAI's o3 model price dropped by 80%

OpenAI's o3 model price dropped by 80%

I woke up to the news that OpenAI had decided to reduce the token price of o3 by 80%. This is great news for everyone interested in AI, and will obviously move the market forward. However, I've got at least 50 articles on this site that should be more interesting than the one I'm writing now. But, unless I either write stuff that's basically "monkey parroting" whatever people tends to be interested in today, or I create content that's of such inflamatory nature that people feel compelled to engaging with, it seems to be impossible for me to "get through". So I am therefor strategically forced into a situation where I've got to publish boring echoes of news that's currently spreading like wild fire through the world to get the attention I think I deserve, and it quite frankly makes me want to throw up.

OpenAI's o3 Model

Personally, I think o3 is overrated. For us it's almost completely irrelevant. We deliver AI chatbots and AI agents to our clients, and GPT-4.1 is more than enough. 4.1 also have API compatibility with the older models, such as GPT 4o, so it implies much less work for us to integrate such models into our tech than o3. I don't think I have ever adviced one of our clients to use o3, and the price reductions is quite frankly irrelevant, even though we do support it in our tech.

So personally I couldn't care less of the fact that OpenAI reduces the price of o3 by 80%. However, since I am proud of what me and Barnabas are doing in AINIRO, and I want others to look at what we've done, and consider it for their next AI project - My hands are tied, and I'm forced into publishing garbage content, about irrelevant news, I personally don't give a f*** about. Because I've got no doubts what so ever that o3's price reduction will be the only thing people will be talking about on LinkedIn today. So if I want to run a successful company, I am forced to writing about things I don't give a sjit about, to such generate the attention I think I deserve.

Innovation

At AINIRO we had MCP 9 months before MCP was even a word. Technologically we were ahead of basically everything that's out there, and every single competitor are looking at our tech with envious eyes, wishing they could do at least half of what we can do. When I wrote about our abilities to execute AI functions from the LLM, it should have been the most important news of that week, easily, hands down. However, as usual we got less than 50 reads on an article that proved we were basically 9 months ahead of everybody, including OpenAI and Anthropic, on technology that's at the core of creating AI agents.

There is a lot of incredibly interesting innovations happening outside of the echo chambed of Silicon Valley. If you doubt this, let me debunk your doubts with one word; DeepSeek! Unfortunately it's almost impossible to be heard unless you've basically sold your soul to YCombinator, or some other corruption-based "startup lubricator". This doesn't mean that YC is better as a partner, it just implies that every single journalist that's currently bubbling up to the top of the SERP, happens to be a prostitute, working for YC or some other "triple AAA" VC fund.

So here we are, o3 had its price reduced by 80%, and I'm forced into trying to squeeze my brain for something intelligent to say about it, such that maybe, just maybe I get lucky, and our website bubbles to the top for a couple of days, bringing us a couple of additional clients, further securing mine and Barnabas' ability to support our families.

Well, I'm sorry to disappoint, I really don't have anything smart to say, because there is nothing smart to say. o3 is a good model for its use cases I guess. And it's great that OpenAI is reducing its price by 80%. And OpenAI being able to continue improving on their product is a good thing I guess. However, it makes me wonder ...

How many really interesting news did I miss today, because of everybody being forced to talking about something, that really nobody actually gives a sjit about ...?

Don't ask me, I've got no idea - Because I am too busy writing boring content about boring news nobody really gives a sjit about ...

Thomas Hansen

Thomas Hansen

I am the CEO and Founder of AINIRO.IO, Ltd. I am a software developer with more than 25 years of experience. I write about Machine Learning, AI, and how to help organizations adopt said technologies. You can follow me on LinkedIn if you want to read more of what I write.

This article was published 11. Jun 2025

The Pragmatic Developer would have used Hyperlambda

On average Hyperlambda will reduce your codebase size by 90 to 96%, depending upon the task. This makes Hyperlambda 10x in productivity.

Everything in Computing is CRUD

CRUD is 'the art of dealing with data'. Fundamentally everything in computing that requires state, is simple CRUD.

Vibe Coding an Edge Function in 30 Seconds

And edge function allows you to securely integrate your own data with the public internet. In Magic you can create such edge functions in seconds using 'vibe coding'.

Copyright © 2023 - 2025 AINIRO.IO Ltd