You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

Martin Krasser 79ed66f870 EXPERIMENTAL Generic LLM wrapper to support chat model interface with configurable chat prompt format (#8295) ... ## Update 2023-09-08 This PR now supports further models in addition to Lllama-2 chat models. See [this comment](#issuecomment-1668988543) for further details. The title of this PR has been updated accordingly. ## Original PR description This PR adds a generic `Llama2Chat` model, a wrapper for LLMs able to serve Llama-2 chat models (like `LlamaCPP`, `HuggingFaceTextGenInference`, ...). It implements `BaseChatModel`, converts a list of chat messages into the [required Llama-2 chat prompt format](https://huggingface.co/blog/llama2#how-to-prompt-llama-2) and forwards the formatted prompt as `str` to the wrapped `LLM`. Usage example: ```python # uses a locally hosted Llama2 chat model llm = HuggingFaceTextGenInference( inference_server_url="http://127.0.0.1:8080/", max_new_tokens=512, top_k=50, temperature=0.1, repetition_penalty=1.03, ) # Wrap llm to support Llama2 chat prompt format. # Resulting model is a chat model model = Llama2Chat(llm=llm) messages = [ SystemMessage(content="You are a helpful assistant."), MessagesPlaceholder(variable_name="chat_history"), HumanMessagePromptTemplate.from_template("{text}"), ] prompt = ChatPromptTemplate.from_messages(messages) memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True) chain = LLMChain(llm=model, prompt=prompt, memory=memory) # use chat model in a conversation # ... ``` Also part of this PR are tests and a demo notebook. - Tag maintainer: @hwchase17 - Twitter handle: `@mrt1nz` --------- Co-authored-by: Erick Friis <erick@langchain.dev>		10 months ago
..
agents	Clean up deprecated agents and update __init__ in experimental (#12231)	11 months ago
autonomous_agents	Type hints on varargs and kwargs that take anything should be `Any`. (#11950)	11 months ago
chat_models	EXPERIMENTAL Generic LLM wrapper to support chat model interface with configurable chat prompt format (#8295)	10 months ago
comprehend_moderation	Fixed some grammatical and Exception types issues (#12015)	11 months ago
cpal	Add CI check that integration tests compile (#12090)	11 months ago
data_anonymizer	Anonymizer small fixes (#11915)	11 months ago
fallacy_removal	adding new chain for logical fallacy removal from model output in chain (#9887)	1 year ago
generative_agents	Use a submodule for pydantic v1 compat (#9371)	1 year ago
graph_transformers	Diffbot Graph Transformer / Neo4j Graph document ingestion (#9979)	1 year ago
llm_bash	Upgrade experimental package dependencies and use Poetry 1.6.1. (#11339)	11 months ago
llm_symbolic_math	Add SymbolicMathChain to experiment in preparation for deprecation (#11129)	12 months ago
llms	LM Format Enforcer Integration + Sample Notebook (#12625)	11 months ago
open_clip	Add Chroma multimodal cookbook (#12952)	10 months ago
openai_assistant	Move OAI assistants to langchain and add callbacks (#13236)	10 months ago
pal_chain	Upgrade experimental package dependencies and use Poetry 1.6.1. (#11339)	11 months ago
plan_and_execute	Use a submodule for pydantic v1 compat (#9371)	1 year ago
prompt_injection_identifier	HF Injection Identifier Refactor	1 year ago
prompts	rename repo namespace to langchain-ai (#11259)	12 months ago
pydantic_v1	`poetry lock` the experimental package. (#9478)	1 year ago
retrievers	add missing init files (#12114)	11 months ago
rl_chain	Revise vowpal_wabbit notebook	12 months ago
smart_llm	Upgrade experimental package dependencies and use Poetry 1.6.1. (#11339)	11 months ago
sql	fix llm_inputs duplication problem in intermediate_steps in SQLDatabaseChain (#10279)	12 months ago
synthetic_data	fix experimental imports (#10875)	1 year ago
tabular_synthetic_data	Install and use `ruff format` instead of black for code formatting. (#12585)	11 months ago
tools	Clean up deprecated agents and update __init__ in experimental (#12231)	11 months ago
tot	Install and use `ruff format` instead of black for code formatting. (#12585)	11 months ago
utilities	Clean up deprecated agents and update __init__ in experimental (#12231)	11 months ago
__init__.py	Add version to langchain_experimental (#11613)	12 months ago
py.typed	Add `py.typed` file to `langchain-experimental`. (#9557)	1 year ago