deepseek/deepseek-v4-flash
The low-cost member of the DeepSeek V4 family. Supports both thinking and non-thinking modes under one id, a native 1M-token context window, and the cheapest per-token rate in the lineup. Use for high-volume, latency-sensitive tasks.
來源：deepseek_2026_08_official_peak · 核驗於 2026-08-20
快取讀取 $0.014/M · 快取寫入 —/M
API 呼叫時用任一 ID 都可以。
deepseek/deepseek-v4-flash把 ONEHOP_KEY 佔位符替換為你的 API Key。 建立 →
from openai import OpenAI
client = OpenAI(
base_url="https://api.onehop.ai/v1",
api_key="<ONEHOP_KEY>",
)
completion = client.chat.completions.create(
model="deepseek/deepseek-v4-flash",
messages=[{"role": "user", "content": "What is the meaning of life?"}],
)
print(completion.choices[0].message.content)最近 30 天 14,296 次請求