Fastino
Fastino is an applied research lab building high-performance, task-optimized small language models (TLMs) that are faster and more secure than general LLMs, capable of running on CPUs and gaming GPUs.
At a Glance
AI Tools by Fastino
(1)
Pioneer
Fine-Tune and Deploy OSS LLMs
Discussions
No discussions yet
Be the first to start a discussion about Fastino
Latest News
Fastino trains AI models on cheap gaming GPUs and just raised $17.5M led by Khosla
Microsoft-Backed Fastino Launches with $7M to Power GPU-Less AI Models
Fastino Launches Pioneer, the First Agent for Fine-tuning and Inference of LLMs
Ash Lewis named to Forbes 30 Under 30 AI Class of 2026
Products & Services
High-performance, small language models optimized for specific enterprise tasks like extraction and reasoning.
A state-of-the-art open-source model for unified entity extraction and classification.
An autonomous agent that automates the fine-tuning and deployment of small language models.
Scalable APIs for accessing task-optimized models for production AI workflows.
Market Position
Fastino competes by offering faster, cheaper, and task-specific alternatives to frontier models like GPT-4, specifically targeting edge and low-compute environments.
Leadership
Founders
Ash Lewis
CEO and Co-founder. Previously founded DevGPT (a developer agent). Ash's experience with the high costs and latencies of using generalist LLMs at DevGPT inspired the launch of Fastino. Named to Forbes 30 Under 30 AI Class of 2026.
George Hurn-Maloney
COO and Co-founder. Previously a General Partner at EnvisionX Capital and a Silicon Valley venture investor.
Executive Team
Ash Lewis
CEO
Founder of DevGPT; Forbes 30 Under 30 Honoree.
George Hurn-Maloney
COO
Former General Partner at EnvisionX Capital.
Board of Directors
Founding Story
Inspired by the limitations of general-purpose LLMs during his time at DevGPT, Ash Lewis co-founded Fastino with George Hurn-Maloney to provide specialized, cost-effective AI models for enterprise tasks.
Business Model
Revenue Model
API usage fees (consumption-based) and enterprise subscription licenses for specialized model infrastructure.
Pricing Tiers
GLiNER models available for community use via Hugging Face.
Tiered usage pricing for high-performance TLMs and specialized enterprise agents.
Target Markets
- Enterprise AI
- Developers
- Infrastructure
- Enterprise data extraction
- Entity classification and structured parsing
- Autonomous AI agents
- Low-latency production workflows
- Orion Technologies