Pranav Nair

Pranav Nair
Occupation	Software entrepreneur
Known for	Founder of Exla

Pranav Nair is a software entrepreneur and the founder of Exla, a technology company that develops a software development kit (SDK) for optimizing and deploying transformer-based artificial intelligence models across a wide range of hardware platforms. Exla is a participant in Y Combinator's Winter 2025 batch.

Career

Nair founded Exla to address the challenge of running large AI models — including large language models (LLMs), vision-language models (VLMs), vision-language-action models (VLAs), and computer vision models — on resource-constrained and diverse hardware. The company's core product is an SDK that aggressively quantizes AI models to reduce their memory footprint by up to 80% while accelerating inference speeds by 3 to 20 times, requiring only a few lines of code to integrate.

Exla supports deployment across a broad spectrum of hardware targets. On the edge computing side, the SDK is compatible with NVIDIA Jetson and Raspberry Pi devices. It also supports NVIDIA consumer and data center GPUs, Apple Silicon and Intel AVX-512 processors, ARM NEON-based CPUs, and mobile devices including iPhones and Android phones. According to the company's published benchmarks, Exla achieves approximately 200 tokens per second on an NVIDIA Jetson (described as a 5x improvement), 35,000 tokens per second on an NVIDIA H100 (7x faster), 30 tokens per second on an iPhone 15 Pro (4x faster), and 75 tokens per second on an Apple M3 Max (4x faster), when running a GPT open-source 120-billion-parameter model.

The SDK allows developers to load models from sources such as Hugging Face and optimize them for specific hardware targets with optional memory budget constraints. In addition to its pre-optimized model library, Exla offers support for custom model optimization and deployment for clients with unique hardware requirements.

Exla also operates a service for spinning up on-demand GPU clusters, available through a separate platform. The company operates in the artificial intelligence, edge computing, and computer vision sectors.