Serverless inference with per-token pricing (pay for output, input, and cached tokens). Training pricing from $0.50 to $10.00 per 1M training tokens depending on model size and fine-tuning type. On-demand deployments at $0.134 to $0.334 per minute ($8 to $20 per hour) for various GPU types.
Individual: $10/month. Team: $15 per user per month. Enterprise: Custom pricing. All plans start with 14-day free trial.
Key features
Guided training runs
Configuration-led training
Custom training logic
Serverless inference
On-demand dedicated deployments
Reserved capacity
OpenAI compatible API
Model library with latest open models
Reinforcement learning support
Multi-region deployments
Connect any MCP client
Space-based access control with owner, editor, and viewer roles
Version history on every memory
Per-memory attribution and provenance tracking
Row-level security in Postgres
Topic-based organization within Spaces
Works with Cursor, Claude Code, Copilot, ChatGPT
Priority support on Team and Enterprise plans
Personal and Shared Spaces
What makes it different
Drop-in replacement for closed-model APIs with cost savings of 50-75%
Instant production deployment from any training checkpoint in seconds
Optimized inference engine for industry-leading throughput and latency
Elastic and global RL inference scaling
Own the complete learning loop for specialized intelligence
Tool-agnostic across mixed AI stacks instead of locked to one vendor
Works across teams without copy-paste or re-explaining context
Person-first design where everyone maintains their own knowledge slice
Single-team pricing with no enterprise minimum
Shared control through role-based permissions and governance