Announcing our $3.5M seed round and a new open source model

Aug 8, 2023
Today, we are excited to announce that Glaive has raised a $3.5M seed round to move towards our goal of democratizing access to AI by enabling the use of small, use-case specific language models trained on synthetic data.
This round is led by Spark Capital with participation from Village Global and Amjad Masad (CEO of Replit).
Why Glaive?
Helping companies get models into prod for the past 2 years we saw some interesting patterns emerge about what people deploy. Most AI applications right now are built on top of LLM APIs, given their ability to generalise. However, to drive widespread adoption of AI, we believe that models need to be commoditised similar to how software is commoditised right now by providing specialist language models, which are consistent and reliable while also being cheaper and faster than general purpose LLMs.
The biggest blocker to having specialist models today is high quality use-case specific data, and this is what we aim to solve with Glaive by enabling the creation of high-quality synthetic datasets for virtually any task with our platform.
Open Source Release
While we focus on providing specialist language models, we also believe that open source models are an integral part of AI progress and this is why we are also releasing a 2.7B parameter model called glaive-function-calling-v1 which has similar function calling abilities as gpt-3.5 and gpt-4 while being significantly smaller.
The model is capable of having multi-turn conversations and intelligently choosing when to execute a function based on the conversation helping the model browse the internet, operate tools and execute tasks for the user.
Few examples from the model-
You can find and use the model on Huggingface-
Data - https://huggingface.co/datasets/glaiveai/glaive-function-calling
Model - https://huggingface.co/glaiveai/glaive-function-calling-v1