Features
Semantic distillation of terminal output (build, test, infra)
Token cost reduction up to 90%
Context pressure warnings before window full
Hot file and active error tracking across sessions
Dependency graph impact alerts on edits
Anti-hallucination guards with factual labels
Engrams for automatic subtask digests
Session health dashboard (omni session --health)
Session handoff between agents
Multi-agent shared memory via SQLite/MCP
Autonomous loop budgets and goals (v0.6.0)
Maker-Checker verification loops
Streaming distillation pipeline (v0.5.8+)
RewindStore for omitted output retrieval
omni learn for automatic filter generation
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
