Features
End-to-end model training inside production harness
Curriculum autoresearch platform for task and reward curation
On-policy reinforcement learning for domain specialization
Pareto-dominant specialist models outperforming frontier models
Reduced serving cost vs. generalist frontier models
State-of-the-art medical models: Kos-1 Lite, Kos-1 Experimental
Env-free RL at 1T parameters using Kimi K2.5
Rubric judge training methodology for grading
Training/inference mismatch correction for sim2real gap
Scalable data generation for practical domains without manual labeling
Laguna XS.2 model (33B params, 3B active, on-device)
Laguna M.1 model (225B params, 23B active, via API)
256K context length for multi-step reasoning
Multi-agent orchestration with planning and tool use
Sandboxed agent execution environments
Developer surfaces: agents, TUI, IDE extensions, binaries
Custom foundation models deployed on-prem or in VPC
Data connectors to repos, databases, data warehouses
Role-based access control for humans and agents
Executive-grade governance and auditability
On-prem, VPC, and workstation deployment (defense only)
Forward Deployed Research Engineers embed with teams
Real-time observability with end-to-end traces
Air-gapped network support
Open-weight model weights available
