OpenAI Slashes o3 Model Prices by 80%: What It Means
Part of: AI Learning Series Here
Subscribe to JorgeTechBits newsletter
This is an update or maybe an addendum to my previous blog: Selecting the Right AI Model for Your Needs (2025 Guide)
Really good news: OpenAI has just reduced the price of their very capable o3 model by a staggering 80%. This means you can now access one of the most advanced reasoning models for a fraction of its previous cost—just $2 per million input tokens and $8 per million output tokens.
How Does o3 Compare to GPT-4o-mini?
I have been using GPT-4o-mini as my default, but for certain things, it’s worth highlighting how o3 stacks up against it.
GPT-4o-mini is celebrated for its speed, efficiency, and cost-effectiveness, delivering much of o3’s intelligence at a significantly lower price point. It excels in fast, straightforward reasoning, especially for complex math and coding problems, and is particularly strong in vision tasks—often outperforming o3 in image analysis and visual perception.
o3, on the other hand, is OpenAI’s flagship for advanced reasoning. It’s designed for deep, multi-step analysis and shines in scenarios where complex, nuanced answers are needed—think business consulting, research, and technical problem-solving. While GPT-4o-mini offers more readable and elegant solutions for some math problems and is generally faster, o3 tends to provide more rigorous, detailed responses, sometimes using tables or more structured outputs.
This table organizes the key points about each model, highlighting their strengths and features.
| Model | Summary | Key Strengths |
|---|---|---|
| GPT-4o-mini | Ideal for quick, high-volume tasks, especially those involving vision or where speed and cost are priorities. | – Fast execution. – Cost-effective for high-volume needs. |
| o3 | Best for tasks demanding deep reasoning, analytical rigor, and complex problem-solving, particularly in technical and business domains. | – Advanced Reasoning: Tackles multi-step, complex analytical tasks. – Coding and Math: Sets benchmarks in programming and mathematical problem-solving. – Scientific and Technical Analysis: Excels in research, business consulting, and creative ideation. – Visual Perception: Analyzes images, charts, and graphics (less than GPT-4o-mini). – Instruction Following: Generates reliable, structured outputs for technical writing and workflow automation. |
| o3 Pro | A premium, more compute-intensive version of o3 designed for the most demanding tasks. | – Enhanced Reliability: Higher accuracy and consistency. – Advanced Features: Supports image inputs, function calling, structured outputs, and a “background mode” for long-running, asynchronous tasks. |
When to use o3 Pro: If your application requires the utmost in reliability, advanced reasoning, or multimodal capabilities (especially combining text and images), or if you’re running critical workloads where accuracy is paramount, o3 Pro is the right choice. For most advanced reasoning and development tasks, the standard o3 model now offers exceptional value and capability at a much lower price.
OpenAI Model Price Comparison Table (June 2025)
The dramatic price drop for o3 makes advanced AI reasoning accessible to more users, while o3 Pro gives power users and enterprises a robust tool for their most challenging AI tasks. This is a major leap forward in AI accessibility and capability
This table should help you quickly compare costs and choose the model for your needs.
Here’s a clear side-by-side comparison of current OpenAI model prices, including o3, o3 Pro, GPT-4o-mini, and other popular options:
| Model | Input (per 1M tokens) | Output (per 1M tokens) | Cached Input (per 1M tokens) | Notes |
|---|---|---|---|---|
| o3 | $2.00 | $8.00 | $0.50 | Flagship reasoning model, 80% price cut |
| o3 Pro | $20.00 | $80.00 | N/A | Premium, high-compute version |
| o3-mini | $1.10 | $4.40 | $0.275 | Cost-efficient reasoning |
| GPT-4o-mini | $0.40 | $1.60 | $0.10 | Fast, affordable, strong vision |
| GPT-4.1 | $2.00 | $8.00 | $0.50 | Advanced, general-purpose |
| GPT-4.1 mini | $0.40 | $1.60 | $0.10 | Balanced speed and intelligence |
| o4-mini | $1.10 | $4.40 | $0.275 | Fast, efficient reasoning |
| GPT-4o | $5.00 | $15.00 | — | Advanced+, multimodal |
Key Takeaways:
- o3 now matches GPT-4.1’s pricing and is much more accessible after the recent 80% price cut.
- GPT-4o-mini and GPT-4.1 mini remain the most affordable options for high-volume or cost-sensitive use cases.
- o3 Pro is positioned for users needing maximum reasoning power and reliability, at a premium price.
- o3-mini and o4-mini offer a middle ground, balancing cost and advanced reasoning performance.