Features
Autonomous ML engineering from single task prompt
Multi-step reasoning with self-correction and iteration
LLM evaluation and benchmarking across models
Dual-LLM automatic prompt optimization loop
Fine-tuning of LLMs and ML models
RAG pipeline building and optimization
Agent swarm creation and coordination
Integration with VS Code, Cursor, Claude Code, OpenVSX
Bring your own LLM (BYOK) support
GPU sandbox for running experiments on your compute
Versioned artifact management and reporting
Synthetic data generation and dataset engineering
Lite mode and Pro mode for different compute profiles
Credit-based usage metering (per month)
Can run autonomously for days
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
