OpenAI announced plans to wind down the availability of GPT-4.5 through its API, effective July 14. The model, which was released in February, will be replaced by GPT-4.1, launched earlier this week. GPT-4.1 is expected to offer comparable or better performance at a significantly lower cost.
Although GPT-4.5, known as Orion, will still be available in the ChatGPT research preview for paying users, its API access will be phased out as OpenAI focuses on future model development.
GPT-4.5, a large AI model that used extensive computing resources, was trained to enhance capabilities like writing and persuasion. Despite its size, it fell short of “frontier-level” industry benchmarks. The model’s high cost, with rates at $75 per million input tokens and $150 per million output tokens, was a key factor in the decision to discontinue its API access.
OpenAI’s decision to prioritize GPT-4.1 comes as part of their strategy to optimize performance while reducing operational costs.
