OpenAi: New O3-mini Model

Today, we’re excited to highlight the launch of OpenAI o3‑mini — the newest, most cost-efficient model in OpenAI’s high-performance reasoning series. Released in June 2025 Node.js Streams Documentation and previewed as early as December 2024, o3‑mini pushes the boundaries of what smaller models can do, especially in domains like science, mathematics, and coding.

What makes this update special for the developer and research community? o3‑mini brings exceptional reasoning performance while maintaining low latency and affordable usage—a perfect balance Node.js Streams Documentation of speed and precision. For those who have worked with OpenAI’s earlier o1‑mini, this new release is an impressive upgrade in both capability and flexibility.

OpenAI o3‑mini is more than just faster and cheaper—it’s production-ready from day one. At OSSTube, we’re particularly excited that this model now supports developer-first features like:

1. Function calling
2. Structured outputs
3. Developer messages

These features open up significant use cases for agents, assistants, and other app logic built on LLMs. o3‑mini also supports streaming, making it even more useful in real-time applications.

Another standout innovation: developers can now select between three levels of reasoning effort—low, medium, and high. This means you can optimize o3‑mini for speed, complexity, or a balanced mode, depending on your application’s need. While o3‑mini doesn’t support vision tasks (stick with OpenAI o1 for that), it excels in fast and intelligent text-based reasoning.

Currently, o3‑mini is available via the Chat Completions API, Assistants API, and Batch API, gradually rolling out to developers in API usage tiers 3–5.

If you’re a ChatGPT user on the Plus, Team, or Pro plan, great news—o3‑mini is available starting today. It will replace o1‑mini in the model picker, but it’s not just a swap. o3‑mini delivers:

Higher rate limits

Lower latency

Improved STEM performance

More responsive code generation

As a bonus, OpenAI has increased the message limits for Plus and Team users—from 50 messages per day (with o1‑mini) to 150 messages per day using o3‑mini.

Pro users will also benefit from unlimited access to both the default and high-effort versions of o3‑mini. The o3‑mini-high mode offers enhanced reasoning depth—perfect for more complex, technical queries, even if it takes slightly longer to generate.

Another exciting feature that OSSTube believes will benefit our user community: o3‑mini now integrates with search to provide up-to-date, linkable results from across the web. This is particularly useful for developers and technical teams who need real-time answers backed by references.

Though this search integration is still in prototype stage, it marks an important shift toward context-aware, current-event-friendly AI—a capability we expect to see expand across all reasoning models in the future.

In a historic move, OpenAI has made o3‑mini accessible to free ChatGPT users for the first time. If you're using ChatGPT’s free version, you can now select "Reason" in the message composer or regenerate responses using o3‑mini.

This means anyone curious about advanced AI capabilities—whether students, hobbyists, or indie devs—can explore what a powerful small model can do without a subscription.

While OpenAI o1 continues to serve as the broad general reasoning model, o3‑mini is a specialized upgrade for users working in technical fields, including engineering, data science, and web development.

In ChatGPT, o3‑mini runs in medium reasoning mode by default, giving you a sweet spot between speed and depth. If your tasks are more complex, the o3‑mini-high option lets the model “think harder”—ideal for advanced coding, logic-heavy problems, and analytical queries.

we see OpenAI o3‑mini as a milestone for lightweight, affordable AI that doesn’t compromise on intelligence. Whether you're building agents, writing code, solving math problems, or deploying educational tools—this is a model that meets both technical depth and production readiness.

We’ll be exploring o3‑mini with live use cases, tutorials, and hands-on integration examples very soon. Stay tuned to OSSTube for deep dives, experiments, and community projects powered by this exciting new model.