Go offline with the Player FM app!
SE Radio 661: Sunil Mallya on Small Language Models
Manage episode 473326519 series 215
Sunil Mallya, co-founder and CTO of Flip AI, discusses small language models with host Brijesh Ammanath. They begin by considering the technical distinctions between SLMs and large language models.
LLMs excel in generating complex outputs across various natural language processing tasks, leveraging extensive training datasets on with massive GPU clusters. However, this capability comes with high computational costs and concerns about efficiency, particularly in applications that are specific to a given enterprise. To address this, many enterprises are turning to SLMs, fine-tuned on domain-specific datasets. The lower computational requirements and memory usage make SLMs suitable for real-time applications. By focusing on specific domains, SLMs can achieve greater accuracy and relevance aligned with specialized terminologies.
The selection of SLMs depends on specific application requirements. Additional influencing factors include the availability of training data, implementation complexity, and adaptability to changing information, allowing organizations to align their choices with operational needs and constraints.
This episode is sponsored by Codegate.
1024 episodes
SE Radio 661: Sunil Mallya on Small Language Models
Software Engineering Radio - the podcast for professional software developers
Manage episode 473326519 series 215
Sunil Mallya, co-founder and CTO of Flip AI, discusses small language models with host Brijesh Ammanath. They begin by considering the technical distinctions between SLMs and large language models.
LLMs excel in generating complex outputs across various natural language processing tasks, leveraging extensive training datasets on with massive GPU clusters. However, this capability comes with high computational costs and concerns about efficiency, particularly in applications that are specific to a given enterprise. To address this, many enterprises are turning to SLMs, fine-tuned on domain-specific datasets. The lower computational requirements and memory usage make SLMs suitable for real-time applications. By focusing on specific domains, SLMs can achieve greater accuracy and relevance aligned with specialized terminologies.
The selection of SLMs depends on specific application requirements. Additional influencing factors include the availability of training data, implementation complexity, and adaptability to changing information, allowing organizations to align their choices with operational needs and constraints.
This episode is sponsored by Codegate.
1024 episodes
All episodes
×

1 SE Radio 667: Ashley Peacock on Cloudflare 1:04:31


1 SE Radio 666: Eran Yahav on the Tabnine AI Coding Assistant 1:02:05


1 SE Radio 665: Malcolm Matalka on Developing in OCaml with Zero Frameworks 56:10


1 SE Radio 664: Emre Baran and Alex Olivier on Stateless Decoupled Authorization Frameworks 51:54


1 SE Radio 663: Tyler Flint on Managing External APIs 52:27


1 SE Radio 662: Vlad Khononov on Balancing Coupling in Software Design 56:19


1 SE Radio 661: Sunil Mallya on Small Language Models 59:28




1 SE Radio 659: Brenden Matthews on Idiomatic Rust 53:54


1 SE Radio 658: Tanya Janca on Secure Coding 1:11:40


1 SE Radio 657: Hong Minhee on ActivityPub and the Fediverse 40:09


1 SE Radio 656: Ivett Ördög on Rewrite versus Refactor 49:43


1 SE Radio 655: Charles Humble on Professional Skills for Software Engineers 55:20


1 SE Radio 654: Chris Patterson on MassTransit and Event-Driven Systems 1:09:06


1 SE Radio 653: Asanka Abeysinghe on Cell-Based Architecture 1:00:06
Welcome to Player FM!
Player FM is scanning the web for high-quality podcasts for you to enjoy right now. It's the best podcast app and works on Android, iPhone, and the web. Signup to sync subscriptions across devices.