BerriAI
LiteLLM acts as an AI Gateway to simplify model access, spend tracking, and fallbacks across over 100 LLMs using a standardized OpenAI format.
Founding Story
Founded in 2023 as part of Y Combinator's W23 batch, BerriAI created LiteLLM to simplify the complexity of managing and translating calls between diverse LLM providers like Azure, OpenAI, and Cohere. The company emerged just when GPT-4 was initially released, recognizing the need for a unified interface to call multiple LLM APIs.
Discussions
No discussions yet
Be the first to start a discussion about BerriAI
Leadership
Founders
Krrish Dholakia
Former project manager at Microsoft. Currently serves as CEO and Co-Founder of BerriAI/LiteLLM.
Ishaan Jaffer
Former software engineer at Coinbase. Currently serves as CTO and Co-Founder of BerriAI/LiteLLM.
Executive Team
Krrish Dholakia
CEO and Co-Founder
Former project manager at Microsoft
Ishaan Jaffer
CTO and Co-Founder
Former software engineer at Coinbase
Business Model
Revenue Model
Freemium model with open-source Community Edition and paid Enterprise Edition. Revenue from enterprise licenses, custom integrations, professional support, and managed hosting.
Pricing Tiers
Includes Docker image, 100+ LLM provider integrations, logging tools (Langfuse, Arize Phoenix, Langsmith, OTEL), virtual keys, budgets, teams, load balancing, and guardrails. Users pay for their own infrastructure (AWS/GCP) to host it. Does not include admin UI for managing teams, SSO, or advanced data retention policies.
Includes all Open Source features plus Enterprise license, governance features, Okta/Google SSO (free for up to 5 users), granular RBAC (Role-Based Access Control), JWT authentication, audit logs, enterprise support, custom SLAs, and corporate-compliant platform tools.
Target Markets
- Gen AI Enablement Teams
- ML Platform Teams
- Software Development companies
- Enterprise organizations
- AI/ML Engineering teams
- DevOps teams
- Gen AI Enablement/ML Platform Teams using Proxy Server as central service for organization-wide LLM access
- Developers building LLM-powered applications with multi-provider support
- Cost management and budget tracking for LLM usage across projects and teams
- Reliability engineering with fallbacks and retries for high availability
- Enterprise teams needing centralized authentication and access control
- Organizations requiring observability and logging for LLM operations
- Netflix
- NASA
- Adobe
- Rocket Money
History & Milestones
Released v1.81.0-stable with Claude Code web search support and 25% CPU usage reduction
Achieved 240M+ Docker pulls and 1B+ requests served
Raised $1.6M in seed funding led by Y Combinator, Gravity Fund, and Pioneer Fund
Released v1.77.3-stable with Priority Quota Reservation
Released v1.78.0-stable introducing MCP Gateway with 70% p99 latency reduction
