Starter plan is free with $30/month compute credits. Team plan is $250/month plus compute. Enterprise plan is custom. Compute is billed per second for CPU, GPU, and memory with no charges for idle time.
Free plan at $0 USD; Pro plan at $5 USD per month.
Key features
Serve LLMs, image, video, and audio models
Isolated environments for coding agents and RL rollouts
Fine-tune and train models on GPUs
Autoscale compute from zero to thousands of GPUs
Distributed storage for models and weights
Global capacity across 20+ clouds
Serverless functions with HTTPS endpoints
Out-of-the-box observability and logs
Entity-graph spreading activation for cross-vocabulary retrieval
Works with any LLM model
No LLM or GPU in retrieval path
Point-in-time and windowed recall
Timestamped memories
Fixed recall cost as memory grows
What makes it different
Custom infrastructure built for AI including container runtime, storage, and scheduler
Sub-second container startup times for GPU workloads
Pay only by the second with no idle charges
Programmatically create millions of concurrent sandboxes
Reaches answers across unrelated documents that share no keywords, where vector search scores zero
Deterministic retrieval with no model in the path, so results never vary between queries
Constant per-query cost regardless of memory size, even at 32,000 words
No infrastructure to run—no vector database, cluster, or GPU needed
Works with any model, so small local models gain the most from better retrieval