black-forest-labs/flux-1-schnell-lora/edit1 點數起 /張· 編輯圖片

FLUX.1 Schnell LoRA 執行 FLUX.1 Schnell,這是蒸餾過、幾步就能出圖的開源 FLUX 模型,可載入最多三個 LoRA。FLUX.1 Dev 的 LoRA 也能在它上面載入,強度 1 時就能看出效果,改圖可以用遮罩替換一件衣服或一個物件。

在 POST /jobs/createTask 把模型名稱填成 black-forest-labs/flux-1-schnell-lora/edit 就能呼叫,1 點數/張起(US$0.01)。新帳戶送 30 點數,可以先在 Playground 免費試用。

  • 生成圖片
  • 編輯圖片
  • 最高 1.5K
  • 9 種畫面比例
  • 可載入自己的 LoRA
模型名稱black-forest-labs/flux-1-schnell-lora/edit
端點POST https://api.popnow.ai/api/v1/jobs/createTask
# 1. Create the task. Keep the Idempotency-Key if you retry after a network error.
export IDEMPOTENCY_KEY="$(uuidgen)"
curl -X POST https://api.popnow.ai/api/v1/jobs/createTask \
  -H "Authorization: Bearer $POPNOW_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $IDEMPOTENCY_KEY" \
  -d '{
  "model": "black-forest-labs/flux-1-schnell-lora/edit",
  "input": {
    "prompt": "A ceramic teapot on a linen cloth, soft window light, muted palette",
    "image_url": "https://media.popnow.ai/influencer/examples/marco.webp",
    "resolution": "1k",
    "output_format": "jpeg"
  }
}'

# 2. Copy data.taskId from the response, then poll until state is succeeded, failed, canceled or expired.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.popnow.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
  -H "Authorization: Bearer $POPNOW_API_KEY"

input 依下方的 Schema 產生,原樣就能通過驗證。把提示詞和連結換成你自己的即可。

價格

選項點數 /張≈ 美元
標準0.9US$0.009

1 點數 = $0.01 · 全部選項及價格:GET /models/black-forest-labs/flux-1-schnell-lora/edit

每個任務會無條件進位到整數點數,所以最低檔跑一次是 1 點數/張(US$0.01)。

限制

解析度
1K · 1.5K
提示詞長度(字元)
5,000
單一上傳上限
10 MB

參數

這些欄位來自我們驗證每個請求所用的同一份 Schema。沒列出的欄位會被拒絕。

  • promptstring必填

    Describe the whole image you want back, not just the change. If a LoRA has a trigger word, put it first.

    範圍
    1 – 5000 chars
  • image_urlstring必填

    作為起點的那張圖。

    範圍
    image/jpeg, image/png, image/webp, image/gif · ≤ 10 MB
  • mask_urlstring選填

    塗出需要改動的區域。

    範圍
    image/jpeg, image/png, image/webp, image/gif · ≤ 10 MB
  • resolutionstring選填

    輸出尺寸。越大越清晰,價格也越高。

    預設值
    "1k"
    可用值
    "1k""1.5k"
  • strengthnumber選填

    結果可以偏離原圖多遠。

    預設值
    0.8
    範圍
    0 – 1
  • lorasarray<object>選填

    載入你自己的 LoRA,鎖定風格或角色。

    預設值
    []
    範圍
    0 – 3 items
    內部欄位
    • loras[].pathstring必填

      Direct URL of a LoRA weights file.

      範圍
      1 – … chars · application/octet-stream
    • loras[].scalenumber必填

      Weight strength from zero to four.

      預設值
      1
      範圍
      0 – 4
  • output_formatstring選填

    結果的檔案格式。

    預設值
    "jpeg"
    可用值
    "jpeg""png""webp"
  • seedinteger選填

    用同一個數字,結果就能重現。

    範圍
    0 – 2147483647
原始 JSON Schema
{
  "additionalProperties": false,
  "properties": {
    "image_url": {
      "contentMediaType": [
        "image/jpeg",
        "image/png",
        "image/webp",
        "image/gif"
      ],
      "description": "Source image. The result keeps its aspect ratio.",
      "format": "uri",
      "type": "string"
    },
    "loras": {
      "default": [],
      "description": "Up to three LoRA weight files, each with its own strength.",
      "items": {
        "additionalProperties": false,
        "properties": {
          "path": {
            "contentMediaType": [
              "application/octet-stream"
            ],
            "description": "Direct URL of a LoRA weights file.",
            "format": "uri",
            "minLength": 1,
            "type": "string"
          },
          "scale": {
            "default": 1,
            "description": "Weight strength from zero to four.",
            "maximum": 4,
            "minimum": 0,
            "type": "number"
          }
        },
        "required": [
          "path",
          "scale"
        ],
        "type": "object"
      },
      "maxItems": 3,
      "type": "array"
    },
    "mask_url": {
      "contentMediaType": [
        "image/jpeg",
        "image/png",
        "image/webp",
        "image/gif"
      ],
      "description": "Optional black-and-white mask the size of the source: white areas are redrawn, black areas are kept.",
      "format": "uri",
      "type": "string"
    },
    "output_format": {
      "default": "jpeg",
      "description": "Container of the returned image.",
      "enum": [
        "jpeg",
        "png",
        "webp"
      ],
      "type": "string"
    },
    "prompt": {
      "description": "Describe the whole image you want back, not just the change. If a LoRA has a trigger word, put it first.",
      "maxLength": 5000,
      "minLength": 1,
      "type": "string"
    },
    "resolution": {
      "default": "1k",
      "description": "Long side of the output: 1k is about 1,000 pixels, 1.5k about 1,500. The source's aspect ratio is kept.",
      "enum": [
        "1k",
        "1.5k"
      ],
      "type": "string"
    },
    "seed": {
      "description": "Random seed for controlled comparisons.",
      "maximum": 2147483647,
      "minimum": 0,
      "type": "integer"
    },
    "strength": {
      "default": 0.8,
      "description": "How far to move away from the source, from 0 to 1.",
      "maximum": 1,
      "minimum": 0,
      "type": "number"
    }
  },
  "required": [
    "image_url",
    "prompt"
  ],
  "type": "object"
}

FLUX.1 Schnell LoRA 的真實成品

Hoodie to fencing whites
提示詞The same South Asian woman with a high ponytail in the same bright fencing hall, now dressed head to toe in fencing whites, a fitted white fencing jacket with the collar open and white knee breeches with long white socks, mask in her hand, mid-stride toward the camera, confident look.
透過 API 送出的同一個請求
{
  "model": "black-forest-labs/flux-1-schnell-lora/edit",
  "input": {
    "prompt": "The same South Asian woman with a high ponytail in the same bright fencing hall, now dressed head to toe in fencing whites, a fitted white fencing jacket with the collar open and white knee breeches with long white socks, mask in her hand, mid-stride toward the camera, confident look.",
    "image_url": "https://media.popnow.ai/influencer/examples/marco.webp",
    "resolution": "1k",
    "output_format": "jpeg",
    "seed": 1009,
    "strength": 0.9
  }
}

這則是從上傳的檔案開始的:把範例連結換成你自己的檔案連結。

回應、輪詢與回呼

所有模型的用法都一樣:先建立任務,再輪詢,或等待回呼。

先報價

cURL
curl -X POST https://api.popnow.ai/api/v1/jobs/quote \
  -H "Authorization: Bearer $POPNOW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "black-forest-labs/flux-1-schnell-lora/edit",
  "input": {
    "prompt": "A ceramic teapot on a linen cloth, soft window light, muted palette",
    "image_url": "https://media.popnow.ai/influencer/examples/marco.webp",
    "resolution": "1k",
    "output_format": "jpeg"
  }
}'

請求內容和 createTask 相同。它回傳預估扣款和最高可能扣款,不會真的執行。

已受理

HTTP
HTTP/1.1 202 Accepted

{
  "code": 200,
  "msg": "success",
  "data": {
    "taskId": "task_01k3m8x9q2z4v7n5p6r8s0t1w3",
    "state": "queued"
  },
  "request_id": "req_example_create"
}

HTTP 202 代表任務已受理,點數先凍結、尚未扣除。最終扣款不會超過凍結額。

查詢狀態

cURL
curl "https://api.popnow.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
  -H "Authorization: Bearer $POPNOW_API_KEY"

每隔幾秒查詢一次並逐步拉長間隔,state 變成 succeeded、failed、canceled 或 expired 時停止。

完成的任務

JSON
{
  "code": 200,
  "msg": "success",
  "data": {
    "taskId": "task_01k3m8x9q2z4v7n5p6r8s0t1w3",
    "model": "black-forest-labs/flux-1-schnell-lora/edit",
    "state": "succeeded",
    "output": {
      "assets": [
        {
          "url": "https://cdn.popnow.ai/tasks/task_01k3m8x9q2z4v7n5p6r8s0t1w3/output.png?signature=FROM_RESPONSE",
          "expiresAt": "2026-10-17T12:20:00Z",
          "mime": "image/png",
          "width": 1024,
          "height": 1024
        }
      ]
    },
    "settled": true
  },
  "request_id": "req_example_record"
}

請在 expiresAt 之前從 url 下載。之後需要新連結,請呼叫 POST /common/download-url。

驗證回呼

Node.js
import { createHash, createHmac, timingSafeEqual } from "node:crypto";
import express from "express";

const app = express();
// express.raw keeps the exact bytes that were signed; express.json would not match.
app.post("/hooks/popnow", express.raw({ type: "application/json" }), (req, res) => {
  const timestamp = req.header("X-Webhook-Timestamp");
  const signature = req.header("X-Webhook-Signature") ?? "";
  const body = JSON.parse(req.body.toString("utf8"));
  const taskId = body.data?.taskId ?? body.task_id;
  const digest = createHash("sha256").update(req.body).digest("hex");
  const expected = createHmac("sha256", process.env.POPNOW_WEBHOOK_SECRET)
    .update(`${taskId}.${timestamp}.${digest}`)
    .digest("base64");
  const ok = expected.length === signature.length && timingSafeEqual(Buffer.from(expected), Buffer.from(signature));
  // Check the signature first, then reject anything older than five minutes.
  if (!ok || Math.abs(Date.now() / 1000 - Number(timestamp)) > 300) return res.status(401).end();
  res.status(200).end(); // answer fast, then fetch the task and store its files
});

設定 callBackUrl 後,我們會把完成的任務送給你。先用原始請求內容驗證簽名,再信任它。

上傳本機檔案

cURL
curl -X POST https://api.popnow.ai/api/v1/common/upload-url \
  -H "Authorization: Bearer $POPNOW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"contentType": "image/png", "bytes": 482113}'

# PUT the file bytes to data.uploadUrl with every header in data.headers, then commit it:
curl -X POST https://api.popnow.ai/api/v1/files/$FILE_ID/commit \
  -H "Authorization: Bearer $POPNOW_API_KEY"
# The commit returns a popnow://f/... reference. Put it in any image, video or
# audio field of the task input instead of a public https URL.

公開的 https 連結可以直接用。本機檔案要先經過 upload-url 和 commit。