langchain

langchain copied to clipboard

Proof of concept only. Do not merge.

andersenchen

hwchase17

Add batching support to all LLMs

2

Currently only done for OpenAI, but should be done for all

hwchase17

Support for other OpenAI Embeddings (text-similarity, code-search)

1

Hello, Could the `OpenAIEmbeddings` class support the other OpenAI embeddings, such as the text-similarity and code-search engines? Currently we can only specify the engine type (ada, babbage, curie, davinci) during...

varunnair18

Caching for LLMChain calls

1

sjwhitmore

https://github.com/yangkevin2/emnlp22-re3-story-generation

hwchase17

To add tests

asamant21

Some agents get stuck in looping behavior. Add a max iteration argument to avoid infinite while loops.

sjwhitmore

If you have a chain that involves agents and that could have a quantitative answer, have a query which allows you to run the chain N times and take some...

sjwhitmore

Add HuggingFacePipeline LLM

4

https://github.com/hwchase17/langchain/issues/354 Add support for running your own HF pipeline locally. This would allow you to get a lot more dynamic with what HF features and models you support since you...

sam-h-bean