Modal Company Profile

Serverless GPU compute and sandboxes, with per-second billing.

Profile

TypeInference platform
HeadquartersNew York, NY
Websitemodal.com

Inference

Modal serves open-weight models through a large public LLM router. Rank is by tokens of other labs' open-weight models in the 7 days to Oct 9, 2026. Modal models and prices.

ProviderOpen-weight modelsRank, 7 daysTop models
Modal5#24GLM 5.3 Flash, Kimi K3, GLM 5.3

Discuss a transaction

Looking to lend into GPU deals like these? Tell us what you finance and we will come back with the deals that fit.

Prefer email? hello@amcompute.com