OpenAI Launches o3-mini: Cheaper Reasoning, Free for First Time

OpenAI rolled out o3-mini this Friday, the newest model in its family of "reasoning" systems, with two selling points that sum up the moment the company finds itself in: it's cheaper, and for the first time in this tier, it's free in ChatGPT. The company had previewed it back in December alongside a more capable system, o3, but is launching it now amid mounting pressure from Chinese rivals like DeepSeek.

"Today's launch marks [...] an important step toward broadening accessibility to advanced AI in service of our mission," an OpenAI spokesperson told TechCrunch. The message isn't accidental: the company is fighting the perception that it's ceding ground in the AI race, all while — according to reports — laying the groundwork for one of the largest funding rounds in history and pushing forward an ambitious data center project.

What a reasoning model is, and why it matters

Unlike most large language models, reasoning models like o3-mini thoroughly fact-check themselves before giving out results. That self-checking helps them avoid some of the pitfalls that normally trip up other systems. The trade-off is time: they take a bit longer to arrive at a solution, but in exchange tend to be more reliable — though not perfect — in domains like physics.

O3-mini is fine-tuned for STEM problems, specifically programming, math, and science. OpenAI claims the model is "largely on par" with the o1 family — o1 and o1-mini — in terms of capabilities, but runs faster and costs less.

According to the company, external testers preferred o3-mini's answers over those from o1-mini more than half the time. In A/B tests, o3-mini made 39% fewer "major mistakes" on "tough real-world questions" compared to o1-mini, produced "clearer" responses, and delivered answers about 24% faster.

Reasoning to order: low, medium, and high

The most practical new feature in o3-mini is that it lets you dial in how much it "thinks." Through the API, developers can choose between three reasoning effort levels — low, medium, and high — so the model works harder or less hard depending on the task and tolerance for latency.

That flexibility translates into a range of performance relative to the previous family. "With low reasoning effort, o3-mini achieves comparable performance with o1-mini, while with medium effort, o3-mini achieves comparable performance with o1," OpenAI writes. "O3-mini with medium reasoning effort matches o1's performance in math, coding and science while delivering faster responses. Meanwhile, with high reasoning effort, o3-mini outperforms both o1-mini and o1."

In ChatGPT, the model is set to medium reasoning effort by default, which the company says provides "a balanced trade-off between speed and accuracy." Paid users can select "o3-mini-high" in the model picker for "higher intelligence" in exchange for slower responses. Whichever version users choose, the model can work with web search to find up-to-date answers with links to relevant sources, though OpenAI cautions that the feature is still a "prototype."

Free in ChatGPT, with tier-based limits

The other headline is access. O3-mini rolled out to all ChatGPT users this Friday, including free users, who can activate it with the new "Reason" button in the chat bar or by asking ChatGPT to "regenerate" a response. It's the first time OpenAI has offered a reasoning model at no cost.

Limits vary by subscription tier:

• ChatGPT Plus and Team: up to 150 queries per day.
• ChatGPT Pro: unlimited access.
• ChatGPT Enterprise and Edu: rolling out in about a week.

On ChatGPT Gov, the company gave no indication.

A price war that already has a name: DeepSeek

The number that sums up the strategy is the price. Via the API, o3-mini is billed at $0.55 per million cached input tokens and $4.40 per million output tokens — a million tokens equates to roughly 750,000 words. That's 63% cheaper than o1-mini.

The figure only makes sense next to the rival. DeepSeek charges $0.14 per million cached input tokens and $2.19 per million output tokens for its R1 reasoning model through its API. In other words, o3-mini is still more expensive than R1 on both counts, but OpenAI has cut prices to land in a competitive range. The word "competitive" is doing a lot of work here: the company has had to lower its prices to move closer to a rival it has itself alleged might have stolen its intellectual property.

The fine print: not the most powerful model

OpenAI is unusually upfront about what o3-mini is not. It's not the company's most powerful model to date, nor does it beat R1 across every benchmark.

O3-mini beats R1 on AIME 2024 — a test that measures how well models understand and respond to complex instructions — but only with high reasoning effort. It also beats R1 on the programming-focused SWE-bench Verified, though by just 0.1 points and, again, only with high reasoning effort. On low reasoning effort, o3-mini lags R1 on GPQA Diamond, which tests PhD-level physics, biology, and chemistry questions.

The edge over its own predecessor is thin in some areas too. On AIME 2024, o3-mini beats o1 by just 0.3 percentage points when set to high reasoning effort. And on GPQA Diamond, o3-mini doesn't even surpass o1's score on high reasoning effort.

The takeaway is that o3-mini's pitch isn't about smashing absolute records, but about the cost-latency-reliability equation: answering well, fast, and cheap. "While o1 remains our broader general-knowledge reasoning model, o3-mini provides a specialized alternative for technical domains requiring precision and speed," OpenAI wrote.

Safety and availability

OpenAI says o3-mini is as "safe" as, or safer than, the o1 family, thanks to red-teaming efforts and its "deliberative alignment" methodology, which has models "think" through the company's safety policy while responding. According to the company, o3-mini significantly outperforms GPT-4o on difficult safety and jailbreak evaluations.

On the API, the model rolled out to select developers this Friday, though it doesn't yet support image analysis.

What changes from here

The move has two concrete implications. For end users, putting free reasoning in ChatGPT lowers the barrier to a capability that was previously reserved for paying customers. For businesses and developers, adjustable effort allows them to calibrate spending: use the low tier for routine tasks and save the high tier for problems that warrant it.

The backdrop is a race over the cost of reasoning that has accelerated with DeepSeek's emergence. OpenAI frames the release as "another step in OpenAI's mission to push the boundaries of cost-effective intelligence." What this launch makes clear is that price, not just raw power, has become the real battleground.