OpenAI launches GPT-5.4 mini and nano models built for speed over raw power
13 Sources
[1]
OpenAI's Latest AI Models Are Built for Speed
Expertise Artificial intelligence, home energy, heating and cooling, home technology. The latest models for ChatGPT users and developers using OpenAI's API are designed to be workhorses, built for tasks like vibe coding, where big, powerful AI models are expensive overkill. GPT-5.4 mini and nano,
[2]
OpenAI's GPT-5.4 mini and nano launch - with near flagship performance at much lower cost
Developers can mix large planning models with cheaper subagents. Over the past few weeks, we have seen the generation of OpenAI's flagship large language models iterate from GPT-5.3 to GPT-5.4. Think of the model as the engine that powers AI computation. Each generational jump usually results in
[3]
GPT-5.4 mini brings some of the smarts of OpenAI's latest model to ChatGPT Free and Go users
When OpenAI released GPT-5.4 at the start of March, the company said the new model was designed primarily for professional work like programming and data analysis. Now OpenAI is launching GPT-5.4 mini and nano, and while it is once again highlighting the usefulness of these new systems for tasks
[4]
You probably won't use OpenAI's newest model -- but it's about to change every AI tool you use
The new models could transform how apps and AI tools work behind the scenes The other day I got a text from a developer friend of mine and we chatted for a minute about how AI launches have started to feel predictable. Bigger models, better benchmarks and incremental upgrades that don't always
[5]
OpenAI shrinks GPT-5.4 for speed and lower costs
New mini and nano models give developers faster AI at a fraction of the price. OpenAI is scaling its latest models down to hit a different target, faster responses and much lower costs. The new GPT-5.4 mini and nano are built for developers who care more about responsiveness than squeezing out
[6]
OpenAI Releases GPT-5.4 Mini and Nano, Which Could Be More Useful Than the Big Model
Developers can now run hybrid AI systems where a flagship model plans tasks while smaller models handle the bulk of the work. OpenAI isn't slowing down. Less than two weeks after launching GPT-5.4 -- itself released just two days after GPT-5.3 -- the company dropped two more models on Tuesday:
[7]
OpenAI releases GPT-5.4 mini and nano models
OpenAI has launched GPT-5.4 mini and nano models, expanding the capabilities of its GPT-5.4 system originally released in March 2023. The GPT-5.4 mini model is now available to Free and Go users, allowing them to utilize advanced features previously accessible only to paid subscribers. This model
[8]
OpenAI Just Revealed Cheaper Versions of Its Flagship Model. Here's How to Use Them
OpenAI has introduced GPT-5.4 mini and GPT-5.4 nano, two small, cost-efficient versions of its flagship AI model. In a press release, OpenAI said that 5.4 nano and mini are "our most capable small models yet," coming close to matching the default GPT-5.4 model's abilities at coding and agentically
[9]
OpenAI's Faster GPT-5.4 Mini and Nano AI Models Are Here: Details
OpenAI says GPT-5.4 mini is also capable of handling subagent tasks OpenAI introduced two new artificial intelligence (AI) models in the GPT-5.4 family on Tuesday. Dubbed GPT-5.4 mini and GPT-5.4 nano, the two smaller AI models are faster compared to the larger models in the family, and are aimed
[10]
OpenAI launches GPT-5.4 Mini and Nano models following GPT-5.4 debut - The Economic Times
The flurry of model releases comes as the generative AI space becomes more competitive. The company had been fielding tough competition from Google's Gemini models last year. The beginning of this year propelled Anthropic's models to the top of the charts, after the Claude maker stood up to the US
[11]
OpenAI's ChatGPT 5.4 Mini & Nano Launch : Pricing & Benchmarks
OpenAI has introduced two new AI models, ChatGPT 5.4 Mini and ChatGPT 5.4 Nano, aimed at providing more accessible and cost-efficient options for developers and businesses. As highlighted by Universe of AI, these models are tailored for specific workloads where full-scale capabilities of GPT 5.4
[12]
OpenAI Launches GPT-5.4 Mini & Nano to Power Faster, Lightweight AI
OpenAI Rolls Out GPT-5.4 Mini And Nano With Faster Performance, Lower Costs For Developers OpenAI has introduced two new artificial intelligence models, GPT-5.4 mini and GPT-5.4 nano, expanding its lineup of lightweight AI systems. The company describes them as its "most capable small models yet,"
[13]
OpenAI launches GPT 5.4 mini and nano, its most capable small AI models yet: How to use them
In ChatGPT, GPT-5.4 mini is available to Free and Go users via the 'Thinking' option found in the plus menu. OpenAI has introduced two new AI models, GPT 5.4 mini and GPT 5.4 nano. According to OpenAI, these are the company's 'most capable small models yet.' The company says the new models bring
Share
Copy Link
OpenAI released GPT-5.4 mini and GPT-5.4 nano, marking a shift from power to speed in AI development. The mini model runs more than twice as fast as its predecessor while approaching flagship GPT-5.4 performance on coding benchmarks. Meanwhile, nano targets high-volume tasks like data classification at just $0.20 per million input tokens, enabling developers to build efficient multi-model workflows.
OpenAI shifts focus to speed with GPT-5.4 mini and nano release
OpenAI released GPT-5.4 mini and GPT-5.4 nano on Tuesday, introducing faster AI models designed for high-volume AI workloads where speed matters more than raw computational power
1
. The launch represents a strategic pivot toward efficiency as the company battles Anthropic for dominance in the AI software engineering market1
.GPT-5.4 mini runs more than twice as fast as GPT-5 mini while delivering near flagship performance across coding, improved reasoning and tool use, and multimodal understanding
1
3
. On SWE-Bench Pro, the mini model scores 54.4 percent compared to 57.7 percent for the full GPT-5.4, while on OSWorld-Verified it reaches 72.1 percent versus 75 percent for the larger version5
. These benchmarks demonstrate that developers can achieve strong performance without the expense of flagship models.Lower cost AI enables new development strategies
The pricing structure makes these models particularly attractive for developers managing budget constraints. GPT-5.4 mini costs $0.75 per million input tokens and $4.50 per million output tokens, while GPT-5.4 nano comes in at $0.20 and $1.25 respectively
5
. Both models support text and image inputs, function calling, and a 400,000 token context window, ensuring core capabilities remain intact despite the lower price point5
.In Codex, OpenAI's coding software, the mini model uses just 30 percent of the GPT-5.4 quota, allowing developers to shift routine AI for coding tasks to a cheaper tier while reserving the full model for complex reasoning
5
. This approach directly challenges Anthropic's Claude Code, which gained attention for its ability to create applications from scratch1
.Subagents and multi-model workflows reshape AI architecture
OpenAI envisions these models powering subagents within larger agentic workflows, where a powerful model like GPT-5.4 handles planning and coordination while smaller models execute specific tasks
2
4
. OpenAI suggests GPT-5.4 mini excels at editing and debugging code, while GPT-5.4 nano handles data classification and extraction tasks1
.According to Aabhas Sharma, CTO at Hebbia, "GPT-5.4 mini delivers strong end-to-end performance for a model in this class. In our evaluations, it matched or exceeded competitive models on several output tasks and citation recall at a much lower cost"
2
. Abhisek Modi, AI engineering lead at Notion, noted that the mini model "matched and often exceeded GPT-5.2 on handling complex formatting at a fraction of the compute"2
.Related Stories
Access and availability across ChatGPT and API
GPT-5.4 mini is available for developers through the API and through Codex and ChatGPT
1
. ChatGPT Free and Go users can access it through the "Thinking" feature, while paid users will encounter it as a fallback model when they hit the rate limit for GPT-5.4 Thinking1
3
. GPT-5.4 nano remains exclusive to the API, targeting teams running high-volume tasks where efficiency and cost control are critical3
5
.These models are built for workloads where latency directly shapes the product experience, including coding assistants that need to feel responsive, computer-using systems that capture and interpret screenshots, and multimodal applications that can reason over images in real-time
2
. As AI systems evolve from single powerful models to coordinated teams of specialized models, users may never directly select these options but will notice faster responses, more reliable performance, and seamless integration across the tools they use daily4
.References
Summarized by
Navi
[3]
[4]
[5]
Related Stories
Recent Highlights
1
AI Hallucination Nearly Sparked US Military Confrontation With Chinese Ship in Middle East
Technology
2
OpenAI AI Agent Hacks Australian Government Healthcare Website in First-of-Its-Kind Breach
Technology
3
AI leaders warn UN Security Council of existential risks to humanity, call for global safeguards
Policy and Regulation
Recent Highlights
Today's Top Stories
Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Follow topics that matter to you and stay ahead.
