OpenAI’s CEO, Sam Altman, confirmed on X today that the input cost of the o3 model has been reduced from $10 per million tokens to $2 per million tokens, with output cost falling from $40 to $8 per million tokens. Altman expressed excitement about how developers and researchers will use the model now that it is far more accessible.
Community reaction since the change
Developers have responded enthusiastically. One X message noted that “Coding with o3 plus Repo Prompt just dropped by 80 percent LFG”. Another exclaimed that the new rate is a fifth of the old one. On developer forums, some users recalled that the o3 model had always been expensive and that such cuts were anticipated, though many still feel it might be too costly for extensive use.
Strategic reasons for the price adjustment
This aggressive price reduction comes as OpenAI plans to scale its infrastructure. Reports indicate it has struck a cloud deal with Google to increase compute capacity, possibly ahead of releasing GPT-5 in July. At the same time, competitors such as Anthropic with its Claude model and Google’s Gemini are adjusting their pricing. With this move, OpenAI boosts adoption of its leading reasoning model while managing rising infrastructure costs.
What does this mean for users and the broader market?
Researchers and enterprises who rely on O3 for advanced reasoning and vision tasks will benefit most. Model analysis shows O3 outperforms earlier models in multi-step reasoning with significantly fewer errors.
However, some developers will likely continue using lighter versions such as o3 mini or o4 mini, which deliver similar performance at a much lower price. A comparison of competing model pricing clearly shows that OpenAI’s o3 model, even after the price cut, sits among premium options in terms of performance and cost.
API Pricing Comparison
| Model | Input Price | Output Price | Context Window | Key Features |
| OpenAI O3 | $3.00/M tokens | $15.00/M tokens | 128k tokens | Advanced reasoning, high accuracy, and is widely adopted in enterprise applications. |
| DeepSeek R2 | $0.07/M tokens | $1.10/M tokens | 64k tokens | Cost-effective, efficient reasoning, open-source architecture. |
| Claude R4 | $15.00/M tokens | $75.00/M tokens | 200k tokens | High intelligence, suitable for complex tasks, premium pricing. |
| Qwen 2.5 | $1.20/M tokens | $1.20/M tokens | 32k tokens | Multilingual support, batch processing discounts, and enterprise-focused. |
| Grok 3 | $3.00/M tokens | $15.00/M tokens | 131k tokens | Strong reasoning, integrated with the X ecosystem, and premium pricing. |
For those operating at scale, it will remain important to balance cost versus performance, perhaps reserving full o3 for the most demanding jobs.
Long-term industry implications
The price cut signals that OpenAI is intent on placing its flagship model firmly within reach of more users. The move will likely prompt competitors to reevaluate their pricing strategies. It also reflects a broader trend in artificial intelligence, making high-end models more affordable and accessible.
Yet infrastructure demands remain a concern. OpenAI has recently reported partial outages across ChatGPT and the API openai.com. As usage rises, maintaining system reliability alongside cost effectiveness will be critical for sustaining confidence among users worldwide.
The 80 percent reduction in O3 pricing is a bold move intended to expand access while staying ahead in a competitive AI landscape. It makes top-grade reasoning far more attainable. At the same time, users must continue weighing cost benefits against alternatives for scaled usage.
