remi/Perplexica
1
0
Fork 0
mirror of https://github.com/ItzCrazyKns/Perplexica.git synced 2025-06-07 18:38:42 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(app): add search API

Browse Source
This commit is contained in:
ItzCrazyKns
2025-03-20 12:29:52 +05:30
parent 048de2cb74
commit 33b895b75e
4 changed files with 230 additions and 63 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
docs/API/SEARCH.md
View File

@ -20,11 +20,11 @@ The API accepts a JSON object in the request body, where you define the focus mo
{
"chatModel": {
"provider": "openai",
"model": "gpt-4o-mini"
"name": "gpt-4o-mini"
},
"embeddingModel": {
"provider": "openai",
"model": "text-embedding-3-large"
"name": "text-embedding-3-large"
},
"optimizationMode": "speed",
"focusMode": "webSearch",
@ -38,18 +38,18 @@ The API accepts a JSON object in the request body, where you define the focus mo
### Request Parameters
- **`chatModel`** (object, optional): Defines the chat model to be used for the query. For model details you can send a GET request at `http://localhost:3001/api/models`. Make sure to use the key value (For example "gpt-4o-mini" instead of the display name "GPT 4 omni mini").
- **`chatModel`** (object, optional): Defines the chat model to be used for the query. For model details you can send a GET request at `http://localhost:3000/api/models`. Make sure to use the key value (For example "gpt-4o-mini" instead of the display name "GPT 4 omni mini").
- `provider`: Specifies the provider for the chat model (e.g., `openai`, `ollama`).
- `model`: The specific model from the chosen provider (e.g., `gpt-4o-mini`).
- `name`: The specific model from the chosen provider (e.g., `gpt-4o-mini`).
- Optional fields for custom OpenAI configuration:
- `customOpenAIBaseURL`: If you’re using a custom OpenAI instance, provide the base URL.
- `customOpenAIKey`: The API key for a custom OpenAI instance.
- **`embeddingModel`** (object, optional): Defines the embedding model for similarity-based searching. For model details you can send a GET request at `http://localhost:3001/api/models`. Make sure to use the key value (For example "text-embedding-3-large" instead of the display name "Text Embedding 3 Large").
- **`embeddingModel`** (object, optional): Defines the embedding model for similarity-based searching. For model details you can send a GET request at `http://localhost:3000/api/models`. Make sure to use the key value (For example "text-embedding-3-large" instead of the display name "Text Embedding 3 Large").
- `provider`: The provider for the embedding model (e.g., `openai`).
- `model`: The specific embedding model (e.g., `text-embedding-3-large`).
- `name`: The specific embedding model (e.g., `text-embedding-3-large`).
- **`focusMode`** (string, required): Specifies which focus mode to use. Available modes: