8.4 KiB
How to Use Large Model API for Translation
How to use multiple ChatGPT-compatible interfaces (or dedicated interfaces) simultaneously?
If you simply have multiple different keys and want to poll them, just separate them with a `|`.However, sometimes you want to use multiple different API addresses/prompts/models/parameters at the same time to compare translation effects. The method is:
Click the "+" button at the bottom right
A window will pop up, select ChatGPT-compatible interface (or dedicated interface), and give it a name. This will copy the current ChatGPT-compatible interface (or dedicated interface) settings and API.
Activate the copied interface and you can make individual settings. The copied interface can run with the original interface, allowing you to use multiple different settings.
model can be selected from a dropdown list, and some interfaces can dynamically fetch the model list based on the API Interface Address and API Key. After filling in these two fields, click the refresh button next to model to obtain the available model list. If the platform does not support the model retrieval interface and the default list does not include the required model, please refer to the official API documentation to manually enter the model.
ChatGPT-Compatible Interfaces
Most large model platforms use ChatGPT-compatible interfaces.
Since there are so many platforms, it's impossible to list them all. For other interfaces not listed, please refer to their documentation to fill in the corresponding parameters.
Foreign Large Model Interfaces
OpenAI
API Interface Address https://api.openai.com/v1
API Key https://platform.openai.com/api-keys
model https://platform.openai.com/docs/models
x.ai
API Interface Address https://api.x.ai/
API Key https://console.x.ai/
groq
API Interface Address https://api.groq.com/openai/v1/chat/completions
API Key https://console.groq.com/keys
model https://console.groq.com/docs/models Fill in Model ID
OpenRouter
API Interface Address https://openrouter.ai/api/v1/chat/completions
API Key https://openrouter.ai/settings/keys
model https://openrouter.ai/docs/models
Mistral AI
API Interface Address https://api.mistral.ai/v1/chat/completions
API Key https://console.mistral.ai/api-keys/
model https://docs.mistral.ai/getting-started/models/
Azure
API Endpoint URL https://{endpoint}.openai.azure.com/openai/deployments/{deployName}/chat/completions?api-version=2023-12-01-preview
Replace {endpoint} and {deployName} with your endpoint and deployName.
DeepInfra
API Endpoint https://api.deepinfra.com/v1/openai/chat/completions
API Key The API Key must be set to empty, otherwise it will result in an error.
Model You can view the currently available models at https://deepinfra.com/chat. As of the time of writing this document, the freely available models are: meta-llama/Meta-Llama-3.1-405B-Instruct meta-llama/Meta-Llama-3.1-70B-Instruct meta-llama/Meta-Llama-3.1-8B-Instruct mistralai/Mixtral-8x22B-Instruct-v0.1 mistralai/Mixtral-8x7B-Instruct-v0.1 microsoft/WizardLM-2-8x22B microsoft/WizardLM-2-7B Qwen/Qwen2.5-72B-Instruct Qwen/Qwen2-72B-Instruct Qwen/Qwen2-7B-Instruct microsoft/Phi-3-medium-4k-instruct google/gemma-2-27b-it openbmb/MiniCPM-Llama3-V-2_5 mistralai/Mistral-7B-Instruct-v0.3 lizpreciatior/lzlv_70b_fp16_hf openchat/openchat_3.5 openchat/openchat-3.6-8b Phind/Phind-CodeLlama-34B-v2 Gryphe/MythoMax-L2-13b cognitivecomputations/dolphin-2.9.1-llama-3-70b
Cerebras
API Endpoint https://api.cerebras.ai/v1/chat/completions
Model Supports llama3.1-8b llama3.1-70b
API Key After selecting a model and sending a message on https://inference.cerebras.ai, you can intercept the request and check the current Headers -> Request Headers -> Authorization value, which is Bearer demo-xxxxhahaha, where demo-xxxxhahaha is the API Key.
Domestic Large Model Interfaces
DeepSeek
API Interface Address https://api.deepseek.com
API Key https://platform.deepseek.com/api_keys
model https://platform.deepseek.com/api-docs/zh-cn/pricing
Alibaba Cloud Bailian Large Model
API Interface Address https://dashscope.aliyuncs.com/compatible-mode/v1
API Key https://bailian.console.aliyun.com/?apiKey=1#/api-key
model https://help.aliyun.com/zh/model-studio/getting-started/models
ByteDance DouBao Large Model
API Interface Address https://ark.cn-beijing.volces.com/api/v3
API Key Create API Key to obtain
model Create Inference Access Point, fill in Access Point instead of Model
Moonshot AI
API Interface Address https://api.moonshot.cn
API Key https://platform.moonshot.cn/console/api-keys
model https://platform.moonshot.cn/docs/intro
Zhipu AI
API Interface Address https://open.bigmodel.cn/api/paas/v4/chat/completions
API Key https://bigmodel.cn/usercenter/apikeys
model https://bigmodel.cn/dev/howuse/model
Lingyiwanwu
API Interface Address https://api.lingyiwanwu.com
API Key https://platform.lingyiwanwu.com/apikeys
model https://platform.lingyiwanwu.com/docs/api-reference#list-models
SiliconFlow
API Interface Address https://api.siliconflow.cn
API Key https://cloud-hk.siliconflow.cn/account/ak
model https://docs.siliconflow.cn/docs/model-names
iFlytek Spark Large Model
API Interface Address https://spark-api-open.xf-yun.com/v1
API Key Refer to the official documentation to obtain APIKey and APISecret, fill in according to the format of APIKey:APISecret
Tencent Hunyuan Large Model
API Interface Address https://api.hunyuan.cloud.tencent.com/v1
API Key Refer to the official documentation
model https://cloud.tencent.com/document/product/1729/97731
Baidu Qianfan Large Model
API Interface Address https://qianfan.baidubce.com/v2
API Key https://cloud.baidu.com/doc/WENXINWORKSHOP/s/Um2wxbaps
model https://cloud.baidu.com/doc/WENXINWORKSHOP/s/Fm2vrveyu
!> For the API Key, please use the Access Key and Secret Key from Baidu AI Cloud IAM to generate a Bearer Token, which should then be entered as the API Key, or directly enter both in the format {Access Key}:{Secret Key} in the API Key field. Note that this is different from the API Key and Secret Key for the old v1 version of Qianfan ModelBuilder; they are not interchangeable.
Dedicated Interfaces for Specific Platforms
Some large model platforms are not fully compatible with the ChatGPT interface, please fill in the parameters to use in the dedicated interface.
gemini
API Interface Address https://generativelanguage.googleapis.com
API Key https://aistudio.google.com/app/apikey
model https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models
claude
API Interface Address https://api.anthropic.com
API Key https://console.anthropic.com/
model https://docs.anthropic.com/en/docs/about-claude/models