Model Gateway
One governed endpoint in your cloud for every model your agents call
Chalk's Model Gateway is an OpenAI compatible endpoint in front of frontier models or your own custom vLLM. It holds the provider credentials, enforces a model allow-list and a token budget on every key, and fails over automatically when a provider is unavailable.
Explore Chalk Model Gateway
One endpoint, any source
Route to frontier providers or your own trained models, through a single API.
Keep data in your VPC
In-cloud support for weights, context, and routing lets you run sensitive workloads entirely within your boundary.
Automatic Fallback
Configure a fallback list per model provider to ensure availability.
Per-Key Limits
Set token budgets, model allowlists, and rate limits per key.
Built-in observability
Every key tracks its token usage and upstream model cost, and every routing decision is traceable.
Runs with your context layer
Routing stays in your cloud, next to the internal data your domain specific models need to run.
Chalk helps us deliver financial products that are more responsive, more personalized, and more secure for millions of users. It's a direct line from infrastructure to impact.
Govern all model traffic from one place
Every model call passes through the gateway, so the provider credential, the allow-list and the budget govern AI workloads from one place instead of scattered across multiple surfaces.
Model Gateway Docs
Keep up with Chalk
What we've been up to and where to find us next.
Most Innovative Companies
See why Fast Company named us on their list.
Register for the upcoming webinar
Learn how to give agents context for evals and production
Raising the Talent Bar
How Chalk filled 40+ roles in one quarter without lowering the bar
Introducing Chalk Notebooks
Read more about our latest product announcement
Stop building your gateway from scratch
Talk to an engineer about routing agent workloads in your cloud.