OpenAI has announced the launch of all-new GPT-5.4 mini and nano AI models, stating that it marks significant improvements in coding workflows and improved speeds at low costs.
In a blog post, OpenAI said that these are its "most capable small models yet", adding that they bring many of the strengths of GPT‑5.4 to faster, more efficient models designed for high-volume workloads. This comes after the release of GPT-5.4, which currently serves as the flagship model from OpenAI used to handle professional and developer workloads.
What To Expect?
The GPT-5.4 Mini model marks significant improvements over GPT-5 Mini in coding, reasoning, tool use, and multimodal understanding. It runs more than 2X faster than the previous model. Also, it approaches the performance of GPT-5.4 on several evaluations, such as SWE-Bench Pro and OSWorld-Verified. It is being considered as suitable enough for applications where both speed and accuracy matter.
The GPT-5.4 Nano model, on the other hand, is the "smallest, cheapest" version of GPT-5.4. It is mainly for tasks where speed and cost matter most. It marks a major upgrade over GPT-5 nano and has been recommended for classification, data extraction, ranking and coding subagents for handling simpler supporting tasks.
According to OpenAI, the two new models have been built for the "kinds of workloads where latency directly shapes the product experience: coding assistants that need to feel responsive, subagents that quickly complete supporting tasks, computer-using systems that capture and interpret screenshots, and multimodal applications that can reason over images in real-time."
Also Read: OpenAI in Talks for $10 Billion Joint Venture With PE Firms
Data shared by OpenAI suggests that GPT-5.4 Mini performs better than GPT-5 Mini at similar latency levels, while coming closer to the GPT-5.4 model in some cases. Further, it states that the two models are especially effective in coding workflows that benefit from fast iteration.
"The models handle targeted edits, codebase navigation, front-end generation, and debugging loops with low latency, making them a strong fit for coding tasks that need to be completed at faster speeds and lower costs," it stated.
Availability
As per the blog post, GPT-5.4 Mini is available in ChatGPT, the API, and Codex. Nano is aimed at developers using the API. In the API, GPT‑5.4 mini has a 400k context window. It costs $0.75 per 1M input tokens as well as $4.50 per 1M output tokens. The same model is available across the Codex app, CLI, IDE extension and web as well.
In ChatGPT, GPT‑5.4 mini is available to Free and Go users through the “Thinking” feature in the + menu. For others, it is available as a rate limit fallback for GPT‑5.4 Thinking. GPT‑5.4 nano, meanwhile, costs $0.20 per 1M input tokens and $1.25 per 1M output tokens in the API.
Essential Business Intelligence, Continuous LIVE TV, Sharp Market Insights, Practical Personal Finance Advice and Latest Stories — On NDTV Profit.
