ayrisdev/crawl4ai
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs: update text in llm-strategies.md to reflect new changes in LlmConfig

Browse Source
This commit is contained in:
Aravind Karnam
2025-03-03 19:24:44 +05:30
parent f14e4a4b67
commit 504207faa6
1 changed files with 18 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
docs/md_v2/extraction/llm-strategies.md
View File

@@ -20,11 +20,17 @@ In some cases, you need to extract **complex or unstructured** information from
## 2. Provider-Agnostic via LiteLLM ## 2. Provider-Agnostic via LiteLLM
You can use LlmConfig, to quickly configure multiple variations of LLMs and experiment with them to find the optimal one for your use case. You can read more about LlmConfig [here](/api/parameters).
```python
llmConfig = LlmConfig(provider="openai/gpt-4o-mini", api_token=os.getenv("OPENAI_API_KEY"))
```
Crawl4AI uses a “provider string” (e.g., `"openai/gpt-4o"`, `"ollama/llama2.0"`, `"aws/titan"`) to identify your LLM. **Any** model that LiteLLM supports is fair game. You just provide: Crawl4AI uses a “provider string” (e.g., `"openai/gpt-4o"`, `"ollama/llama2.0"`, `"aws/titan"`) to identify your LLM. **Any** model that LiteLLM supports is fair game. You just provide:
- **`provider`**: The `<provider>/<model_name>` identifier (e.g., `"openai/gpt-4"`, `"ollama/llama2"`, `"huggingface/google-flan"`, etc.). - **`provider`**: The `<provider>/<model_name>` identifier (e.g., `"openai/gpt-4"`, `"ollama/llama2"`, `"huggingface/google-flan"`, etc.).
- **`api_token`**: If needed (for OpenAI, HuggingFace, etc.); local models or Ollama might not require it. - **`api_token`**: If needed (for OpenAI, HuggingFace, etc.); local models or Ollama might not require it.
- **`api_base`** (optional): If your provider has a custom endpoint. - **`base_url`** (optional): If your provider has a custom endpoint.
This means you **aren’t locked** into a single LLM vendor. Switch or experiment easily. This means you **aren’t locked** into a single LLM vendor. Switch or experiment easily.
@@ -52,20 +58,19 @@ For structured data, `"schema"` is recommended. You provide `schema=YourPydantic
Below is an overview of important LLM extraction parameters. All are typically set inside `LLMExtractionStrategy(...)`. You then put that strategy in your `CrawlerRunConfig(..., extraction_strategy=...)`. Below is an overview of important LLM extraction parameters. All are typically set inside `LLMExtractionStrategy(...)`. You then put that strategy in your `CrawlerRunConfig(..., extraction_strategy=...)`.
1. **`provider`** (str): e.g., `"openai/gpt-4"`, `"ollama/llama2"`. 1. **`llmConfig`** (LlmConfig): e.g., `"openai/gpt-4"`, `"ollama/llama2"`.
2. **`api_token`** (str): The API key or token for that model. May not be needed for local models. 2. **`schema`** (dict): A JSON schema describing the fields you want. Usually generated by `YourModel.model_json_schema()`.
3. **`schema`** (dict): A JSON schema describing the fields you want. Usually generated by `YourModel.model_json_schema()`. 3. **`extraction_type`** (str): `"schema"` or `"block"`.
4. **`extraction_type`** (str): `"schema"` or `"block"`. 4. **`instruction`** (str): Prompt text telling the LLM what you want extracted. E.g., “Extract these fields as a JSON array.”
5. **`instruction`** (str): Prompt text telling the LLM what you want extracted. E.g., “Extract these fields as a JSON array.” 5. **`chunk_token_threshold`** (int): Maximum tokens per chunk. If your content is huge, you can break it up for the LLM.
6. **`chunk_token_threshold`** (int): Maximum tokens per chunk. If your content is huge, you can break it up for the LLM. 6. **`overlap_rate`** (float): Overlap ratio between adjacent chunks. E.g., `0.1` means 10% of each chunk is repeated to preserve context continuity.
7. **`overlap_rate`** (float): Overlap ratio between adjacent chunks. E.g., `0.1` means 10% of each chunk is repeated to preserve context continuity. 7. **`apply_chunking`** (bool): Set `True` to chunk automatically. If you want a single pass, set `False`.
8. **`apply_chunking`** (bool): Set `True` to chunk automatically. If you want a single pass, set `False`. 8. **`input_format`** (str): Determines **which** crawler result is passed to the LLM. Options include:
9. **`input_format`** (str): Determines **which** crawler result is passed to the LLM. Options include:
- `"markdown"`: The raw markdown (default). - `"markdown"`: The raw markdown (default).
- `"fit_markdown"`: The filtered “fit” markdown if you used a content filter. - `"fit_markdown"`: The filtered “fit” markdown if you used a content filter.
- `"html"`: The cleaned or raw HTML. - `"html"`: The cleaned or raw HTML.
10. **`extra_args`** (dict): Additional LLM parameters like `temperature`, `max_tokens`, `top_p`, etc. 9. **`extra_args`** (dict): Additional LLM parameters like `temperature`, `max_tokens`, `top_p`, etc.
11. **`show_usage()`**: A method you can call to print out usage info (token usage per chunk, total cost if known). 10. **`show_usage()`**: A method you can call to print out usage info (token usage per chunk, total cost if known).
**Example**: **Example**:
@@ -233,8 +238,7 @@ class KnowledgeGraph(BaseModel):
async def main(): async def main():
# LLM extraction strategy # LLM extraction strategy
llm_strat = LLMExtractionStrategy( llm_strat = LLMExtractionStrategy(
provider="openai/gpt-4", llmConfig = LlmConfig(provider="openai/gpt-4", api_token=os.getenv('OPENAI_API_KEY')),
api_token=os.getenv('OPENAI_API_KEY'),
schema=KnowledgeGraph.schema_json(), schema=KnowledgeGraph.schema_json(),
extraction_type="schema", extraction_type="schema",
instruction="Extract entities and relationships from the content. Return valid JSON.", instruction="Extract entities and relationships from the content. Return valid JSON.",