FLUX.1 Dev ControlNet LoRA API 调用参考
FLUX.1 上用姿势、深度或线稿控制，可挂你的 LoRA
用一个 PopNow API 密钥，就能在自己的代码里调用 FLUX.1 Dev ControlNet LoRA。下面列出要填的调用名、这个型号接受的每个输入字段、每次请求花多少积分，以及一段可以直接粘进终端运行的请求。
参数说明直接取自型号的输入 schema，暂时只有英文。
一览
- 出品方
- Black Forest Labs
- 产出
- 图片
- 起价
- 5.4 积分 每张图 ($0.054)
1 积分 = 0.01 美元。任务受理时先冻结可能的最高费用，完成后按实际用量结算、向上取整到整积分，多冻的退回；失败的任务不扣积分。想在运行前知道确切数字，先调 POST /api/v1/jobs/quote。
生成图片
black-forest-labs/flux-1-dev-controlnet-lora/text-to-image
价格
| 选项 | 积分 (每张图) | 美元 |
|---|---|---|
默认 | 5.4 | $0.054 |
参数
| 字段 | 类型 | 默认值 | 取值范围 | 作用 |
|---|---|---|---|---|
prompt必填 | string | — | 最多 5000 个字符 | Describe the picture you want: subject, clothing, setting and light. The control image only sets the layout. |
control_image_url必填 | string (uri) | — | — | Control image: a pose skeleton, depth map, canny or soft-edge line map, or a grayscale or plain photo. The result keeps its aspect ratio. |
control_strength可选 | number | 0.7 | 0 到 1 | How strongly the result follows the control image, from 0 to 1. |
control_guidance_start可选 | number | 0 | 0 到 1 | Fraction of the denoising steps at which the control image starts to apply, from 0 to 1. |
control_guidance_end可选 | number | 0.8 | 0 到 1 | Fraction of the denoising steps at which the control image stops applying, from 0 to 1; not before the start. |
resolution可选 | string | 1k | 1k, 1.5k | Length of the output's long side: 1k is 1024 pixels, 1.5k is 1536. The control image's aspect ratio is kept. |
num_outputs可选 | integer | 1 | 1 到 4 | How many images to generate in one run, from 1 to 4. |
loras可选 | array<object> | [] | 最多 3 项 | Up to three LoRA weight files trained for FLUX.1 dev, each with its own strength. |
guidance_scale可选 | number | 3.5 | 0 到 20 | How closely the image follows the prompt, from 0 to 20. |
steps可选 | integer | 28 | 1 到 50 | Denoising steps, from 1 to 50. Fewer steps return faster and a little softer. |
output_format可选 | string | jpeg | jpeg, png, webp | Container of the returned image. |
seed可选 | integer | — | 0 到 2147483647 | Random seed for controlled comparisons. |
请求示例
这段请求原样就能运行，没写的字段都用默认值。运行前先把密钥放进环境变量 POPNOW_API_KEY。
示例里的 control_image_url 用的是 PopNow 的示例素材，换成你自己的公开 HTTPS 链接、上传文件或已保存的素材即可。
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: $(uuidgen)" \
-d '{
"model": "black-forest-labs/flux-1-dev-controlnet-lora/text-to-image",
"input": {
"prompt": "A cozy reading nook beside a rain-streaked window, warm lamp light, a sleeping orange cat on a knitted blanket, soft film grain",
"control_image_url": "https://media.popnow.ai/influencer/examples/marco.webp",
"num_outputs": 1
}
}'响应里会立刻带回 taskId。之后用 GET /api/v1/jobs/recordInfo 取结果，或者在请求里加上 callBackUrl 等通知。
curl "https://api.popnow.ai/api/v1/jobs/recordInfo?taskId=TASK_ID_FROM_CREATE_TASK" \
-H "Authorization: Bearer $POPNOW_API_KEY"