LLM Stats
To ensure AI is built on truth by creating the largest collection of high-quality evaluations across all AI modalities (language, vision, code, reasoning, etc.).
Founding Story
Founded with the goal of providing transparent, independent benchmarking for the rapidly evolving LLM landscape and building self-improving AI agents.
Discussions
No discussions yet
Be the first to start a discussion about LLM Stats
Leadership
Founders
Jonathan Chavez
Co-founder of ZeroEval and LLM Stats. Background in RL & physics research. Won 1st place in the Imperial College London AI Hackathon out of 200 teams worldwide.
Sebastian Crossa
Co-founder of ZeroEval. Previously a founding engineer at Micro (backed by venture capital). Software entrepreneur educated at Tecnológico de Monterrey.
Executive Team
Jonathan Chavez
Co-Founder
Research background in RL and Physics; hackathon champion.
Sebastian Crossa
Co-Founder
Former founding engineer at Micro; experienced software developer.
Business Model
Revenue Model
Likely usage-based API fees for LLM access and evaluation infrastructure (SaaS).
Pricing Tiers
Usage-based pricing for model tokens through ZeroEval's unified API.
Target Markets
- AI Research Labs
- Enterprise Generative AI Teams
- AI Developer Community
- LLM benchmarking and comparison
- AI agent quality optimization
- RL research and testing
- Enterprise AI model selection
- Fortune 100 companies
History & Milestones
ZeroEval was founded by Jonathan Chavez and Sebastian Crossa in New York.
Accepted into the Y Combinator Summer 2025 (S25) batch.
Launched the LLM-Stats leaderboard and reached 200,000+ monthly active users.
Introduced 'Autotune' and auto-optimization tools for AI agents.
