LangChain

These are the classes to use with LangChain.

class imagine.langchain.ImagineChat(*args, client, async_client, api_key=None, endpoint=None, max_retries=5, timeout=120, verify=False, name=None, cache=None, verbose=None, callbacks=None, tags=None, metadata=None, custom_get_token_ids=None, callback_manager=None, rate_limiter=None, disable_streaming=False, model='Llama-3.1-8B', temperature=0.0, max_tokens=None, top_k=None, top_p=None, streaming=False, frequency_penalty=None, presence_penalty=None, repetition_penalty=None, stop=None, max_seconds=None, ignore_eos=None, skip_special_tokens=None, stop_token_ids=None)

A chat model that use Imagine Inference API

classmethod get_lc_namespace()

Get the namespace of the langchain object.

classmethod is_lc_serializable()

Return whether this model can be serialized by Langchain.

class imagine.langchain.ImagineLLM(*args, client, async_client, api_key=None, endpoint=None, max_retries=5, timeout=120, verify=False, name=None, cache=None, verbose=None, callbacks=None, tags=None, metadata=None, custom_get_token_ids=None, callback_manager=None, model='Llama-3.1-8B', temperature=0.0, max_tokens=None, top_k=None, top_p=None, streaming=False, frequency_penalty=None, presence_penalty=None, repetition_penalty=None, stop=None, max_seconds=None, ignore_eos=None, skip_special_tokens=None, stop_token_ids=None)

An LLM that uses Imagine Inference API

class imagine.langchain.ImagineEmbeddings(*, client, async_client, api_key=None, endpoint=None, max_retries=5, timeout=120, verify=False, max_concurrent_requests=64, model='BAAI/bge-large-en-v1.5')

Imagine embedding models.

async aembed_documents(texts)

Embed a list of document texts.

Args:

texts: The list of texts to embed.

Returns:

List of embeddings, one for each text.

async aembed_query(text)

Embed a single query text.

Args:

text: The text to embed.

Returns:

Embedding for the text.

embed_documents(texts)

Embed a list of document texts.

Args:

texts: The list of texts to embed.

Returns:

List of embeddings, one for each text.

embed_query(text)

Embed a single query text.

Args:

text: The text to embed.

Returns:

Embedding for the text.