LangChain recently raised a hefty amount of funding, with the Python framework securing millions at a soaring $200 million valuation. This comes after the same project, dedicated to LLMs, raised seed funding at a much smaller amount. With its focus on abstraction wrappers that make it easy for developers to integrate LLMs into their programs, the open-source platform has become well-known for the convenience it brings. This can be seen with its over 20,000 stars on GitHub, yet questions still linger about the inflated valuation of the project.
Developed by Harrison Chase, a programmer, LangChain was launched at the crest of the generative AI wave in October 2022. Initially, the main offering of LangChain was purely text-based and only supported OpenAI and Cohere APIs, as well as a Python interpreter. However, as time has gone by the framework has evolved, becoming more compatible with a range of model providers and hosting platforms. At this time, the platform has over 50 document loaders, more than 10 vector databases, and over 15 tools commonly used by LLMs.
LangChain has been likened to and compared to Pandas and NumPy, the popular machine learning libraries. Similarly, LangChain is said to have brought significant usability improvements to LLMs. This is due to the framework being able to chain commands together and even provide built-in benchmarks to evaluate a Model’s potential utility. Moreover, LangChain also introduces memory to LLMs, empowering programs by connecting them to an LLM, or leveraging a large dataset by connecting an LLM to it.
However, despite its utility and features, the truth is that LangChain is a somewhat unfavourable choice for use in a production environment. With many developers swearing against using the framework due to its anachronisms, the platform has not been able to get out of user’s bad books. Karrot_Kream, a member of the Hacker News forum, commented on its poor performance, saying “I tell friends who ask about LangChain that it’s great to experiment with but not something I’d put into production.”
The arrival of instant messaging bots changed the outlook of LLMs, and increased the demand for OpenAI’s APIs that aided the development of these programs. Now, with open source APIs like OpenAI’s, developers find the ease in connecting GPT models with programs. Furthermore, since LangChain encapsulates models from different platforms, its underlying value still remains, potentially becoming the “Hugging Face of LLMs” in the future.
LangChain was founded by Harrison Chase, a programmer and tech industry veteran who has a keen interest in the intersection of artificial intelligence and human language. Chase holds multiple patents in the area of natural language processing and has a wealth of experience in developing and scaling AI applications. His capabilities have been recognised by some of the industry’s leading players, leading to investors pumping millions of dollars into LangeChain. Prior to his focus on LangChain, he has been integral in the development of large-scale projects, such as an investment tracking platform used within the financial services sector.
In conclusion, while the value still remains, it may be a little exaggerated judging by the current utility of LangChain. With APIs such as OpenAI’s now competing with it, it can be difficult to compete in terms of matters such as performance and observability. However, with Chase’s expertise, the platform’s current momentum, and ongoing investment, it is likely that the framework will indeed grow into the beast which it is projected to become. With pre-build services on offer, and an increased focus on open source models, LangChain may just become the iOS of OpenAI’s Android.
