Nano Banana 系列開發指南

{
  "contents": [{
    "role": "user",
    "parts": [
      { "fileData": { "fileUri": "https://your-cdn.example.com/a.png", "mimeType": "image/png" } },
      { "fileData": { "fileUri": "https://your-cdn.example.com/b.jpg", "mimeType": "image/jpeg" } },
      { "text": "把第二張圖裡的外套穿到第一張圖的人物身上" }
    ]
  }],
  "generationConfig": {
    "responseModalities": ["IMAGE"],
    "imageConfig": { "imageSize": "2K" }
  }
}

"promptTokensDetails": [
  { "modality": "TEXT", "tokenCount": 17 },
  { "modality": "IMAGE", "tokenCount": 1120 }
]

curl --location 'https://api.apiyi.com/v1beta/models/gemini-3-pro-image:generateContent' \
  --header "Authorization: Bearer $APIYI_API_KEY" \
  --header 'Content-Type: application/json' \
  --data '{
      "contents": [{
          "role": "user",
          "parts": [
              {
                  "fileData": {
                      "fileUri": "https://raw.githubusercontent.com/apiyi-api/ai-api-code-samples/refs/heads/main/Vision-API-OpenAI/otter.png",
                      "mimeType": "image/png"
                  }
              },
              { "text": "add five dogs" }
          ]
      }],
      "generationConfig": {
          "responseModalities": ["IMAGE"],
          "imageConfig": { "aspectRatio": "16:9", "imageSize": "2K" }
      }
  }' > output.json

import base64, os, requests

API_KEY = os.environ["APIYI_API_KEY"]
MODEL = "gemini-3.1-flash-image"
URL = f"https://api.apiyi.com/v1beta/models/{MODEL}:generateContent"

IMAGE_URLS = [
    "https://raw.githubusercontent.com/apiyi-api/ai-pics/refs/heads/main/1762260696217_dd0352c1f9604540.png",
]
PROMPT = "把人物的衣服換成藍色夾克，其餘保持不變"

payload = {
    "contents": [{
        "role": "user",
        "parts": [
            *[{"fileData": {"fileUri": u, "mimeType": "image/png"}} for u in IMAGE_URLS],
            {"text": PROMPT},
        ],
    }],
    "generationConfig": {
        "responseModalities": ["IMAGE"],
        "imageConfig": {"imageSize": "2K"},
    },
}

resp = requests.post(URL, json=payload, timeout=300,
                     headers={"Authorization": f"Bearer {API_KEY}"})
if resp.status_code != 200:
    raise SystemExit(f"HTTP {resp.status_code}: {resp.text[:500]}")
data = resp.json()

# 1. 確認參考圖真的被讀到了：promptTokensDetails 裡要有 IMAGE
details = data.get("usageMetadata", {}).get("promptTokensDetails", [])
image_tokens = sum(d.get("tokenCount", 0) for d in details if d.get("modality") == "IMAGE")
if image_tokens == 0:
    print("⚠️ 參考圖沒有被讀到（谷歌沒拉到圖），這張結果不會參考原圖")

# 2. 遍歷 parts 取圖，不要按下標
for part in data["candidates"][0]["content"]["parts"]:
    inline = part.get("inlineData")
    if inline:
        with open("output.png", "wb") as f:
            f.write(base64.b64decode(inline["data"]))
        print(f"✅ 已儲存 output.png（參考圖 {image_tokens} tokens）")
        break
else:
    print("響應中沒有圖片：", data["candidates"][0].get("finishReason"))

{
  "contents": [{ "parts": [{ "text": "畫一張東京今天天氣的卡片海報" }] }],
  "tools": [{ "googleSearch": {} }]
}

API Connection Error: HTTPSConnectionPool(host='api.apiyi.com', port=443): Read timed out. (read timeout=120)

timeout = {
    "1K": 300,  # 5 分鐘 - 快速預覽
    "2K": 300,  # 5 分鐘 - 推薦使用
    "4K": 600,  # 10 分鐘 - 超高畫質
}

import requests, base64

API_KEY = "sk-your-api-key"
URL = "https://api.apiyi.com/v1beta/models/gemini-3.1-flash-image:generateContent"
H = {"Authorization": f"Bearer {API_KEY}", "Content-Type": "application/json"}
CFG = {"responseModalities": ["IMAGE"], "imageConfig": {"imageSize": "2K"}}

contents = []  # 全程維護同一個對話歷史

def turn(instruction, save_to):
    contents.append({"role": "user", "parts": [{"text": instruction}]})
    data = requests.post(URL, headers=H,
                         json={"contents": contents, "generationConfig": CFG}, timeout=300).json()
    part = next(p for p in data["candidates"][0]["content"]["parts"] if "inlineData" in p)
    contents.append({"role": "model", "parts": [part]})   # 關鍵：把產出圖回填進歷史
    with open(save_to, "wb") as f:
        f.write(base64.b64decode(part["inlineData"]["data"]))

turn("生成一隻橙色的貓，坐在藍色沙發上，簡筆畫風格", "step1.png")
turn("把沙發改成紅色，貓和構圖保持不變", "step2.png")     # 基於上一輪的圖
turn("給貓戴一頂黃色小帽子，其它保持不變", "step3.png")     # 繼續累積，紅沙發會保留

cand = (resp.get("candidates") or [{}])[0]
parts = (cand.get("content") or {}).get("parts") or []      # 相容 parts=null
images = [p["inlineData"] for p in parts if "inlineData" in p]
if not images:
    raise RuntimeError(f"未返回圖片，finishReason={cand.get('finishReason')}")

final = images[-1]                                 # 多圖時最後一張為最終稿
image_bytes = base64.b64decode(final["data"])
mime = final["mimeType"]                           # 以響應為準，別寫死 image/png

const parts = resp?.candidates?.[0]?.content?.parts ?? [];
const images = parts.filter((p) => p.inlineData?.data);   // ✅ 不要寫 parts[1]
if (images.length === 0) throw new Error("Gemini 未返回圖片資料");
const { data, mimeType } = images[images.length - 1].inlineData;   // 最後一張為最終稿

parts = response["candidates"][0]["content"]["parts"] or []   # 安全拒絕時 parts 為 null
images = [p["inlineData"]["data"] for p in parts if "inlineData" in p]

if images:
    final_image = images[-1]   # 最後一張 = 最終稿