diff --git a/README.md b/README.md index 7fadd94..8630314 100644 --- a/README.md +++ b/README.md @@ -25,6 +25,7 @@ Stability AI Luma Photon Luma Photon + Ideogram V_2 ## 1. Introduction @@ -56,6 +57,9 @@ - `Baidu ERNIE irag-1.0` - `Stable Diffusion 3.5 large turbo` - `Luma Photon` + - `Ideogram V_2` + + 项目架构流程如下： ```mermaid @@ -239,6 +243,12 @@ MLLM 模型主要用于自动切片后的切片标题生成，此功能默认关 请自行[注册账号](https://lumalabs.ai/api/keys)并申请 API Key，填写到 `bilive.toml` 文件中对应的 `LUMA_API_KEY` 中。 +##### 3.3.7 Ideogram V_2 模型 + +> 如需使用 Ideogram V_2 模型，请将 `IMAGE_GEN_MODEL` 参数设置为 `ideogram`。 + +请自行[注册账号](https://ideogram.ai/manage-api)并申请 API Key，填写到 `bilive.toml` 文件中对应的 `IDEOGRAM_API_KEY` 中。 + #### 4. bilitool 登录 > 由于一般日志打印不出二维码效果（docker 的日志不确定是否能打印，等发布新image时再修改，docker 版本请先参考文档 [bilive](https://bilive.timerring.com)，本 README 只针对源码部署），所以这步需要提前在机器上安装 [bilitool](https://github.com/timerring/bilitool): diff --git a/assets/ideogram.svg b/assets/ideogram.svg new file mode 100644 index 0000000..ba6f7f6 --- /dev/null +++ b/assets/ideogram.svg @@ -0,0 +1 @@ +Ideogram \ No newline at end of file diff --git a/bilive.toml b/bilive.toml index 8992de9..2292541 100644 --- a/bilive.toml +++ b/bilive.toml @@ -37,7 +37,7 @@ qwen_api_key = "" # Apply for your own Qwen API key at https://bailian.console.a [cover] generate_cover = false # whether to generate cover -image_gen_model = "minimax" # the image generation model, can be "minimax" or "siliconflow" or "tencent" or "baidu" or "stability" or "luma" +image_gen_model = "minimax" # the image generation model, can be "minimax" or "siliconflow" or "tencent" or "baidu" or "stability" or "luma" or "ideogram" minimax_api_key = "" # Apply for your own Minimax API key at https://platform.minimaxi.com/user-center/basic-information/interface-key siliconflow_api_key = "" # Apply for your own SiliconFlow API key at https://cloud.siliconflow.cn/i/3Szr5BVg tencent_secret_id = "" # Apply for your own Tencent Cloud API key at https://console.cloud.tencent.com/cam/capi @@ -45,3 +45,4 @@ tencent_secret_key = "" # Apply for your own Tencent Cloud secret key as above baidu_api_key = "" # Apply for your own Baidu API key at https://console.bce.baidu.com/iam/key/list stability_api_key = "" # Apply for your own Stability API key at https://platform.stability.ai/account/keys luma_api_key = "" # Apply for your own Luma API key at https://lumalabs.ai/api/keys +ideogram_api_key = "" # Apply for your own Ideogram API key at https://ideogram.ai/manage-api diff --git a/src/config.py b/src/config.py index 532b833..569d569 100644 --- a/src/config.py +++ b/src/config.py @@ -80,4 +80,5 @@ TENCENT_SECRET_ID = config.get('cover', {}).get('tencent_secret_id') TENCENT_SECRET_KEY = config.get('cover', {}).get('tencent_secret_key') BAIDU_API_KEY = config.get('cover', {}).get('baidu_api_key') STABILITY_API_KEY = config.get('cover', {}).get('stability_api_key') -LUMA_API_KEY = config.get('cover', {}).get('luma_api_key') \ No newline at end of file +LUMA_API_KEY = config.get('cover', {}).get('luma_api_key') +IDEOGRAM_API_KEY = config.get('cover', {}).get('ideogram_api_key') \ No newline at end of file diff --git a/src/cover/cover_generator.py b/src/cover/cover_generator.py index 4dc6c1e..45b7b21 100644 --- a/src/cover/cover_generator.py +++ b/src/cover/cover_generator.py @@ -77,6 +77,10 @@ def cover_generator(model_type): from .image_model_sdk.luma_sdk import luma_generate_cover return luma_generate_cover(cover_path) + elif model_type == "ideogram": + from .image_model_sdk.ideogram_sdk import ideogram_generate_cover + + return ideogram_generate_cover(cover_path) else: upload_log.error(f"Unsupported model type: {model_type}") return None diff --git a/src/cover/image_model_sdk/ideogram_sdk.py b/src/cover/image_model_sdk/ideogram_sdk.py new file mode 100644 index 0000000..28bafae --- /dev/null +++ b/src/cover/image_model_sdk/ideogram_sdk.py @@ -0,0 +1,51 @@ +import requests +import json +import os +import time +from src.config import IDEOGRAM_API_KEY + + +def ideogram_generate_cover(your_file_path): + """Generater cover image using ideogram V_2 model + Args: + your_file_path: str, path to the image file + Returns: + str, local download path of the generated cover image file + """ + try: + url = "https://api.ideogram.ai/remix" + + files = {"image_file": open(your_file_path, "rb")} + payload = { + "image_request": json.dumps( + { + "prompt": "This is a video screenshot, please generate a cover in the style of a manga", + "aspect_ratio": "ASPECT_10_16", + "image_weight": 75, + "magic_prompt_option": "ON", + "model": "V_2", + } + ) + } + headers = {"Api-Key": f"{IDEOGRAM_API_KEY}"} + + response = requests.post(url, data=payload, files=files, headers=headers) + if response.status_code == 200: + response_json = response.json() + image_url = response_json["data"][0]["url"] + img_data = requests.get(image_url).content + cover_name = time.strftime("%Y%m%d%H%M%S") + ".png" + temp_cover_path = os.path.join(os.path.dirname(your_file_path), cover_name) + with open(temp_cover_path, "wb") as handler: + handler.write(img_data) + os.remove(your_file_path) + return temp_cover_path + else: + raise Exception(response.text) + except Exception as e: + print(e, flush=True) + return None + + +if __name__ == "__main__": + print(ideogram_generate_cover(""))