docs(rerankers): align reranker config docs with v3 SDK source
This commit is contained in:
@@ -61,11 +61,13 @@ All rerankers share these common configuration parameters:
|
||||
|
||||
### LLM Reranker
|
||||
|
||||
| Parameter | Description | Type | Default |
|
||||
| -------------- | --------------------------- | ------ | -------- |
|
||||
| `llm.provider` | LLM provider for reranking | `str` | Required |
|
||||
| `llm.config` | LLM configuration object | `dict` | Required |
|
||||
| `top_n` | Number of results to return | `int` | `None` |
|
||||
| Parameter | Description | Type | Default |
|
||||
| ----------- | -------------------------------------------------------------------------------------------- | ------ | -------------- |
|
||||
| `provider` | LLM provider (`openai`, `anthropic`, etc.) | `str` | `"openai"` |
|
||||
| `model` | LLM model to use for scoring | `str` | `"gpt-4o-mini"` |
|
||||
| `api_key` | API key for LLM provider | `str` | `None` |
|
||||
| `top_k` | Number of results to return after reranking | `int` | `None` |
|
||||
| `llm` | Optional nested LLM config (`{"provider": ..., "config": {...}}`) for provider-specific fields (e.g. `ollama_base_url`) | `dict` | `None` |
|
||||
|
||||
## Environment Variables
|
||||
|
||||
|
||||
Reference in New Issue
Block a user