Publicado em 2026-08-16 · 赵予安
Resposta direta
OpenAI GPT-4.1 is listed at 0.289 input per million tokens, with a 1,047,576 token context window. CacheSafety for this exact id is unpublished. Este guia foi escrito para equipes de produto e plataforma que comparam qualidade de modelos, custo, política de roteamento e risco de rollout.
What shipped
OpenAI GPT-4.1 is now in the NextModel gateway catalog under the public slug openai--gpt-4.1. Its catalog model id is openai/gpt-4.1. The display name is "OpenAI: GPT-4.1". The provider is OpenAI. The model has a context window of 1,047,576 tokens. Its capability list has streaming, tool, JSON, vision, and long-context support. The latency band is 1000 to 3000 ms. The catalog entry comes from the NextModel gateway catalog with a Go origin.
Pricing
Input price is $0.289 per 1M tokens. Output price is $1.157 per 1M tokens. For comparison, OpenAI: o3 lists the same $0.289 input price and the same $1.157 output price. OpenAI: GPT-5.6 Terra has the same input price but an output price of $1.736 per 1M tokens. OpenAI: GPT-5.2 is $0.253 input and $2.025 output. GPT-4.1 has the lowest output price among these related entries.
Who this is for
The catalog lists vision, long-context, and agent as the use cases. This is for teams that pass images, process very long documents, or run tool-driven agent workflows. The one to three second latency band fits interactive API calls. It is a long-context model with vision input in one request.
What is still unverified
The release date is unpublished in this catalog snapshot. Independent third-party quality scores are not in these notes. There is no first-party CacheSafety Bench row for this exact model. Safe Hit, Bad Hit, and trap rates are unpublished.