OpenAI is not the pioneer in bringing generative AI to businesses – Cohere, based out of Toronto with ties to Google, already offers such services to enterprises. Instead, OpenAI has lately declared a facility that follows up on this concept by releasing their consumer product ChatGPT for organizations.
Martin Kon, President and COO at Cohere, discussed how the firm’s machine learning systems are employed within larger organizations.
Cohere, boasting an enviable legacy because of two founders from renowned Google Brain, reignited the ongoing enthusiasm around AI-generated systems as its young age (~few years) testifies.
Google Brain introduced the ‘transformer’ model for Natural Language Processing in 2017. From there, Aidan Gomez (CEO) and Nick Frosst (CTO) at Cohere joined forces with Ivan Zhang to bring this NLP to the commercial sphere.
Bringing the power of Google to Cohere, Martin Kon joined them last month to lead the business operations. After spending six years at YouTube, this was a perfect opportunity for him to utilize his vast expertise, which is currently paying dividends as the company is booming.
API calls made to Cohere have increased significantly every month for the last year – growth of 65% – and there has also been a corresponding rise in the number of developers utilizing Cohere’s services.
Cohere’s prime focus is now on providing its expansive language models and related software to businesses. With it gaining recognition, this move is aimed at furnishing value to corporate entities.
Kon, by contrast, contending that its approach is distinct from OpenAI’s, maintains that it is taking a fundamentally different route to this problem.
Kon says:
“We’re working with developers in organizations, the AI/ML teams, to bring these capabilities into their organizations.”
“OpenAI wants you to bring your data to their models, exclusive to Azure. Cohere wants to bring our models to your data, in whatever environment you feel comfortable in.”
Technical Comparison Of Cohere Vs. Openai: Which Is Better?
ChatGPT is a type of Large Language Model (LLM) offered by Cohere, focusing on generation rather than representation. Representation-typeOn the other hand, representation-type LLMs are used for understanding language like sentiment analysis.
The size of each type of model varies and can be classified as small, medium, large, and large. The model’s size has pros and cons, with speed performance decreasing or increasing depending on the selected size.
Stanford HELM: Holistic Evaluation of Language Models – gives an objective ranking of Cohere’s base model, clocking in at 52 billion parameters.
Citing Stanford’s HELM website, the largest version of Cohere’s model, known as “xlarge,” involves 175B parameters. OpenAI’s GPT-3 DaVinci model is similar to its largest version and includes the same number of parameters.
The competition between Cohere and OpenAI in the enterprise is one of speed, innovation, and engagement. CIOs face a difficult decision when choosing which platform to use for their organizations.
While Cohere can offer rapidly updated solutions for bottom-line customer experience and data-driven insights, OpenAI provides engaging AI-powered support that can quickly be adapted to customer needs. Both provide powerful tools to enhance organizational performance and customer satisfaction.
Source: thenewstack.io
