OpenRouter, the routing layer that gives developers access to hundreds of AI models through a single API, closed a $113 million Series B on May 28. CapitalG (Alphabet's independent growth fund) led the round. NVIDIA Ventures, ServiceNow Ventures, MongoDB Ventures, Snowflake Ventures, and Databricks Ventures joined as new investors, alongside returning backers Andreessen Horowitz and Menlo Ventures.

What Happened

OpenRouter now routes requests to 400+ models spanning text, image, audio, speech, transcription, embedding, and video. Weekly token volume climbed from 5 trillion to 25 trillion over the past six months, a 5x increase, putting the platform on pace to exceed one quadrillion tokens annually. The platform serves more than 8 million developers. The investor lineup signals enterprise momentum: NVIDIA builds the GPU hardware these models run on, while Snowflake, MongoDB, Databricks, and ServiceNow are where enterprise data lives.

Why It Matters for Creators

Managing API access to a dozen AI providers meant a dozen API keys, a dozen billing dashboards, and a dozen different failure modes to handle in code. OpenRouter collapses all of that to a single endpoint. When a provider is slow or unavailable, the platform routes automatically to an alternative without requiring changes to your workflow or pipeline code.

The NVIDIA Ventures investment points toward tighter integration with NVIDIA's NIM inference infrastructure. For creators running heavy workloads (image generation, video synthesis, audio models), lower-latency GPU-optimized routes are the likely outcome as that partnership deepens.

Key Details

  • Round: $113M Series B
  • Lead investor: CapitalG (Alphabet's independent growth fund)
  • New investors: NVentures (NVIDIA), ServiceNow Ventures, MongoDB Ventures, Snowflake Ventures, Databricks Ventures
  • Returning investors: Andreessen Horowitz, Menlo Ventures
  • Weekly token volume: 25 trillion (up from 5T six months ago, 5x growth)
  • Annual run rate: On pace for 1 quadrillion tokens
  • Developers served: 8 million+
  • Models: 400+ across text, image, audio, video, and embedding modalities

How to Integrate

OpenRouter uses OpenAI-compatible API endpoints. To route through it from any tool that uses the OpenAI SDK, set your base URL to https://openrouter.ai/api/v1 and swap in your OpenRouter API key. Claude, GPT-5.5, Gemini, FLUX, and hundreds of other models become available through the same interface. For image generation pipelines currently calling providers directly, the endpoint swap adds automatic failover at no extra cost. OpenRouter earns on volume, not markup over provider pricing.

What to Do Next

If you're managing separate API keys across multiple AI providers, OpenRouter is worth evaluating as a consolidation layer. Enterprise controls include workspaces, spend management, provider-level logging policies, and zero-data-retention options for studios handling sensitive creative assets. Visit openrouter.ai to compare provider pricing across models and review the model catalog before deciding which routes to use in production.