PRICING
Competitive pricing. Billed per token.
Simple Pricing
Built for Scale
Competitive, per-token pricing.
Set the latency target on every request. Volume pricing is available for custom and fine-tuned models.
Pay only for the tokens you serve. Choose a route optimized for latency or cost, with volume pricing for custom and fine-tuned models.
What you get
Vanilla and fine-tuned OSS models
Request-level latency targets
0 GPUs sitting idle
What you get
Vanilla and fine-tuned OSS models
Request-level latency targets
0 GPUs sitting idle
Get Started
Ship the model. We’ll scale the inference.
Request access to Terminal for open-source and custom model serving.
Get Started
See Why Top Finance Teams Use Lateral
Request access to Terminal for open-source and custom model serving.
Get Started
Ship the model. We’ll scale the inference.
Request access to Terminal for open-source and custom model serving.