GPT-5.4 Nano (OpenAI)

GPT-5.4 Nano is positioned as the cost-efficient member of OpenAI’s GPT-5.4 series, intended for high-volume, latency-sensitive applications such as real-time chat, automated data processing, and agent workflows. A third-party launch analysis describes the series as a hybrid approach that combines larger-model planning with smaller-model execution, although that architecture claim is not independently documented here.

The model shows a practical balance across general reasoning, coding, multimodal, and agent-oriented evaluations: it scores strongly on LiveCodeBench, AIME, GPQA Diamond, LegalBench, and long-context reasoning, while its Terminal-Bench Hard result indicates room for improvement in difficult terminal-based software tasks. Its strongest practical fit is therefore broad, efficient task handling and agent support, especially where higher reasoning effort can be used selectively.

LLM Gatewayopenai/gpt-5.4-nanogpt-nano

Open provider page Provider docs

Quick Info

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

Cost

Input token cost: $0.20
Output token cost: $1.25

Limits

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

Transparent token rates

Compare GPT-5.4 Nano (OpenAI) 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 Nano (OpenAI)

LLM Gateway

CoverageAnalysis

An Altioric.ai analysis dated March 17, 2026 covers the OpenAI GPT-5.4 series launch and explicitly names GPT-5.4 nano as OpenAI's most cost-efficient offering to date, priced at one-fifth of the flagship model's cost. The article reports series-wide improvements of approximately 2x faster inference and an 80% cost red The piece positions GPT-5.4 nano as a production-ready, price-disruptive tier of the GPT-5.4 family, arguing that the combination of 2x speed gains and 80% cost reductions makes previously cost-prohibitive AI-agent applications economically sustainable. Marketing-leaning framing (e.g., "revolutionary," "transformationa

LLM Gateway

CoverageBenchmark

A third-party benchmark aggregator page explicitly titled for OpenAI GPT-5.4 Nano lists Nano pricing at $0.20 per 1M input tokens and $1.25 per 1M output tokens, positioning it as OpenAI's lowest-cost offering in the GPT-5.4 series. The page tracks 79 benchmark rows specifically for GPT-5.4 Nano across categories inclu Published GPT-5.4 Nano benchmark scores include a 42.4% success rate on Terminal-Bench Hard (92nd percentile), 27.4% Pass@1 on Tau3-Banking (79th percentile), an Elo score of 1,104 on GDPval-AA (77th percentile), and 76.0% success on Tau2-Bench Telecom (71st percentile), with measurement dates ranging from June to Sept