GPT 5.4 Mini

GPT-5.4 Mini is OpenAI's compact successor to GPT-5 mini, designed as a high-throughput alternative that retains the core character of the larger GPT-5.4 family while cutting latency and cost. It is positioned as a fast-lane model that trades some raw capability for substantially quicker responses and lower per-token pricing, making it attractive for production deployments where speed and economics matter as much as peak accuracy. Sources describe it as roughly twice as fast as its predecessor, signaling that OpenAI prioritized inference efficiency alongside the usual quality gains expected from a generational refresh.

In practice, GPT-5.4 Mini is well suited to chat interfaces, coding assistants, and agent workflows that run at scale, where reliable instruction following, multi-step reasoning, and consistent tool use carry the workload. It accepts text and image inputs and exposes the modern API feature set, including tool and function calling, structured outputs, and web search, which lets developers build agentic applications without sacrificing the production-grade tooling around the model. The combination of a large context window, multimodal input, and a lower price point makes it a flexible middle ground for teams that need more than a nano-tier model but want to avoid the cost of the full GPT-5.4.

Vercel AI Gatewayopenai/gpt-5.4-minigpt-mini

Open provider page Provider docs

Quick Info

Provider: Vercel AI Gateway
Model key: openai/gpt-5.4-mini
Release date: Mar 17, 2026
Last updated: Mar 17, 2026
Knowledge cutoff: 2025-08-31
Input modalities
Output modalities
Capabilities

Cost

Input token cost: $0.75
Output token cost: $4.50

Limits

Input tokens: 272,000 tokens
Output tokens: 128,000 tokens
Context window: 400,000 tokens

Transparent token rates

Compare GPT 5.4 Mini pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about GPT 5.4 Mini

Vercel AI Gateway

Official sourceAnnouncement

A Vercel Community announcement by Vercel Alum Pauline P. Narvas, dated March 17, 2026, confirms that OpenAI's GPT-5.4 Mini and GPT-5.4 Nano have been added to Vercel AI Gateway and frames them as built for speed and efficiency in complex tasks like coding and computer use. The post highlights three developer-facing po To start using these models, the post instructs developers to set their model identifier to openai/gpt-5.4-mini or openai/gpt-5.4-nano in the AI SDK. The thread closes automatically in two months and currently shows 59 views with one related link, indicating modest but documented engagement on the gateway rollout.

Vercel AI Gateway

Coverage

Tech News News: OpenAI has launched two new AI models — GPT-5.4 mini and GPT-5.4 nano — aimed at delivering faster performance and lower costs for high-volume workloa.

Vercel AI Gateway

Coverage

OpenAI unveils compact GPT-5.4 mini and nano models, delivering faster responses and lower costs for coding and multimodal workloads.

Vercel AI Gateway

CoverageRelease Notes

OpenAI is releasing today two new small models, GPT-5.4 mini and nano, which are optimized for speed and efficiency.

Vercel AI Gateway

CoverageBenchmark

Independent Stratix evaluation of GPT-5.4, Mini, and Nano. MATH-500, AIME 2025, HLE, and SWE-bench results across all three variants.

Vercel AI Gateway

Official sourceRelease Notes

Vercel's changelog dated March 17, 2026 announced that GPT-5.4 Mini and GPT-5.4 Nano from OpenAI are now available on Vercel AI Gateway, with both models described as delivering state-of-the-art performance for their size class in coding and computer use and as built for sub-agent workflows where multiple smaller model For Mini specifically, Vercel positions it as handling code generation, tool orchestration, and multi-step browser interactions more reliably than previous mini-tier models and recommends it as a strong default for agentic tasks that need to balance capability and cost. Developers invoke the model by setting model to o

Vercel AI Gateway

CoverageRelease Notes

EdTech Innovation Hub's March 19, 2026 report confirms the GPT-5.4 mini and GPT-5.4 nano launch, framing the release as OpenAI's response to growing demand for production AI systems that balance performance, cost, and latency. The article notes the announcement was made via an OpenAI for Business LinkedIn post and unde The piece reiterates OpenAI's positioning that GPT-5.4 mini runs more than twice as fast as GPT-5 mini while approaching GPT-5.4's performance on benchmarks like SWE-Bench Pro and OSWorld-Verified, and that GPT-5.4 nano is positioned as the smallest, cheapest variant for classification, data extraction, ranking, and su