Update docs/API_REFERENCE.md
Browse files- docs/API_REFERENCE.md +68 -0
docs/API_REFERENCE.md
CHANGED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# API Reference
|
| 2 |
+
|
| 3 |
+
This document describes the public Python modules and functions available in AnyCoder.
|
| 4 |
+
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
## `models.py`
|
| 8 |
+
|
| 9 |
+
### `ModelInfo` dataclass
|
| 10 |
+
|
| 11 |
+
```python
|
| 12 |
+
@dataclass
|
| 13 |
+
class ModelInfo:
|
| 14 |
+
name: str
|
| 15 |
+
id: str
|
| 16 |
+
description: str
|
| 17 |
+
default_provider: str = "auto"
|
| 18 |
+
```
|
| 19 |
+
|
| 20 |
+
### `AVAILABLE_MODELS: List[ModelInfo]`
|
| 21 |
+
|
| 22 |
+
A list of supported models with metadata.
|
| 23 |
+
|
| 24 |
+
### `find_model(identifier: str) -> Optional[ModelInfo]`
|
| 25 |
+
|
| 26 |
+
Lookup a model by name or ID. Returns a `ModelInfo` or `None`.
|
| 27 |
+
|
| 28 |
+
---
|
| 29 |
+
|
| 30 |
+
## `inference.py`
|
| 31 |
+
|
| 32 |
+
### `chat_completion(model_id: str, messages: List[Dict[str,str]], provider: Optional[str]=None, max_tokens: int=4096) -> str`
|
| 33 |
+
|
| 34 |
+
Send a one-shot chat completion request. Returns the assistant response as a string.
|
| 35 |
+
|
| 36 |
+
### `stream_chat_completion(model_id: str, messages: List[Dict[str,str]], provider: Optional[str]=None, max_tokens: int=4096) -> Generator[str]`
|
| 37 |
+
|
| 38 |
+
Stream partial generation results, yielding content chunks.
|
| 39 |
+
|
| 40 |
+
---
|
| 41 |
+
|
| 42 |
+
## `hf_client.py`
|
| 43 |
+
|
| 44 |
+
### `get_inference_client(model_id: str, provider: str="auto") -> InferenceClient`
|
| 45 |
+
|
| 46 |
+
Create and return a configured `InferenceClient`, routing to Groq, OpenAI, Gemini, Fireworks, or HF as needed.
|
| 47 |
+
|
| 48 |
+
---
|
| 49 |
+
|
| 50 |
+
## `deploy.py`
|
| 51 |
+
|
| 52 |
+
### `send_to_sandbox(code: str) -> str`
|
| 53 |
+
|
| 54 |
+
Wrap HTML code in a sandboxed iframe via a data URI for live preview.
|
| 55 |
+
|
| 56 |
+
### `load_project_from_url(url: str) -> Tuple[str, str]`
|
| 57 |
+
|
| 58 |
+
Import a Hugging Face Space by URL, returning status message and code content.
|
| 59 |
+
|
| 60 |
+
---
|
| 61 |
+
|
| 62 |
+
## `plugins.py`
|
| 63 |
+
|
| 64 |
+
### `PluginManager`
|
| 65 |
+
|
| 66 |
+
* `discover()`: auto-discovers plugins in the `plugins/` namespace.
|
| 67 |
+
* `list_plugins() -> List[str]`: return registered plugin names.
|
| 68 |
+
* `run_plugin(name: str, payload: Dict) -> Any`: execute a plugin action.
|