DeepInfra for startups
DeepInfra is a serverless AI inference platform providing ultra-fast, cost-efficient access to the world's leading open-source models including Llama, Mistral, and Qwen. Pay-per-token pricing with no minimums and sub-100ms inference latency.
DeepInfra Startup Program Benefits and Tiers
Available offer
Free AI inference tokens with DeepInfra DeepStart
DeepInfra DeepStart gives qualifying early-stage AI companies up to 1 billion free inference tokens — the cost-effective ML inference platform for running state-of-the-art open-source models including Llama; Mistral; Qwen; and Stable Diffusion in production.
Eligibility
Early-stage AI startups building with open-source models; new DeepInfra customers
- Access
- Available to eligible PerkBook startups
