openai/gpt-5.4-mini
The efficient member of the GPT-5 family, tuned for high-throughput, latency-sensitive workloads. Text + image input, 400K-token context, strong reasoning/coding/tool use at the lowest GPT-5 per-token rate.
出典: launch_0023 · 確認日 2026-06-01
Fast $0.15 / $0.9 /M
キャッシュ読み取り $0.0075/M · キャッシュ書き込み —/M
| ティア | 入力 | キャッシュ読み取り | キャッシュ書き込み | 出力 |
|---|---|---|---|---|
| Standard | $0.075 | $0.0075 | — | $0.45 |
| Priority | $0.15 | $0.015 | — | $0.9 |
どちらの ID を使っても API でこのモデルを呼び出せます。
openai/gpt-5.4-miniONEHOP_KEY プレースホルダーをご自身の API Key に置き換えてください。 作成する →
from openai import OpenAI
client = OpenAI(
base_url="https://api.onehop.ai/v1",
api_key="<ONEHOP_KEY>",
)
completion = client.chat.completions.create(
model="openai/gpt-5.4-mini",
messages=[{"role": "user", "content": "What is the meaning of life?"}],
)
print(completion.choices[0].message.content)OpenAI's flagship GPT-5.6 for the hardest reasoning, coding, and agentic work.
$0.25/M↓ · $1.5/M↑
Balanced GPT-5.6 quality and cost for production workloads.
$0.125/M↓ · $0.75/M↑
The most efficient GPT-5.6 for high-volume and latency-sensitive work.
$0.05/M↓ · $0.3/M↑
OpenAI GPT-5.5 — flagship, 1M context, top reasoning & agentic coding.
$0.5/M↓ · $3/M↑
OpenAI GPT-5.4 — strong general model, 1M context, lower cost than 5.5.
$0.25/M↓ · $1.5/M↑
過去 30 日間で 4,046 件のリクエスト