게시일 2026-08-16 · 赵予安
핵심
OpenAI GPT-5.4 is in the NextModel catalog at 2.50 / 15 per million tokens, with a 1,050,000 token context window. CacheSafety for this exact id is unpublished. 이 가이드는 모델 품질, 비용, 라우팅 정책, 운영 출시 리스크를 비교하는 프로덕트 및 플랫폼 팀을 위해 작성되었습니다.
What shipped
OpenAI GPT-5.4 is listed in the NextModel live catalog with model id openai/gpt-5.4 and public slug openai--gpt-5.4. The display name is OpenAI: GPT-5.4, and the provider is OpenAI. The catalog entry has capabilities for streaming, tool use, JSON output, vision, and long context. Latency band is 1000 to 3000 ms.
Pricing and context
Input price is $2.50 per 1 million tokens. Output price is $15 per 1 million tokens. The context window is 1,050,000 tokens. Related catalog entries have different prices. OpenAI GPT-4o is $2.50 in and $10 out. OpenAI GPT-4.1 is $2 in and $8 out. OpenAI GPT-5.6 Terra is $2 in and $12 out.
Who this is for
The catalog lists use cases as vision, long-context, and agent.
Unverified
Release date is unpublished in this catalog snapshot. Independent third-party quality scores are unpublished in the supplied notes. The NextModel bench table has no first-party CacheSafety row for this exact model. Safe Hit, Bad Hit, and trap rates are unpublished.