Pricing
Simple, usage-based pricing.
Start free with the Auto model. Scale up to open models, GPUs, and frontier models like Claude and GPT as your work grows, and only pay for what you use.
Free
$0/mo
Get started with the Auto model on a CPU sandbox.
Get started- Auto model for fast, low-cost everyday work
- CPU sandbox with 8 vCPU and 8 GB RAM
- No credit card required
Plus
$30/mo
Open models and a GPU sandbox for heavier work, with a monthly usage allowance and optional pay-as-you-go.
Start with Plus- Open models like DeepSeek V4 Pro and Kimi K2.6
- Entry GPUs (T4, L4 and A10) up to 24 GB VRAM
- Up to 32 GB sandbox RAM
- Optional pay-as-you-go past your allowance
Pro
Popular$90/mo
Every model, including Claude and GPT, plus the A100 GPU and a larger monthly usage allowance.
Start with Pro- Every model, including Claude (Sonnet, Opus) and GPT
- Mid-tier GPUs (L40S and A100) up to 80 GB VRAM
- Up to 64 GB sandbox RAM
- At least 4× the monthly usage allowance of Plus
- Optional pay-as-you-go past your allowance
Max
$200/mo
Everything unthrottled, with all models, all GPUs, the highest limits and the largest usage allowance.
Start with Max- All models and every GPU at the highest limits
- Top-end GPUs (H100 and H200) up to 141 GB VRAM
- Up to 128 GB sandbox RAM
- At least 12× the monthly usage allowance of Plus
- Priority capacity and pay-as-you-go
Enterprise
Need higher limits, dedicated capacity, SSO, or custom billing for your team? Book a call and we'll design a plan around your workload.