This is not prompt engineering. Stratus operates in learned representation space — encoding state, simulating futures, and selecting actions before a single token is sent to your LLM.
How It Works
State arrives
Raw environment observations — DOM state, tool outputs, structured data — are passed to Stratus alongside a goal description. Nothing is assumed. Everything is encoded fresh.
Encoder compresses
The State Encoder converts both the current state and the goal into compact learned representations. These embeddings capture what actually matters for task completion — not surface text, but semantic meaning your agent can reason over.
World Model simulates
Before any action is taken, the World Model asks: “what happens if I do X?” It simulates the predicted next state for each candidate action — in representation space, in milliseconds, without touching the real environment.
Planner sequences
The Planning Layer runs a forward search through those simulations — selecting and sequencing actions until the predicted outcome converges on the goal. It returns a ranked action sequence, the predicted state at each step, and a confidence score.
LLM executes with context
The verified plan is injected into an enriched execution prompt and forwarded to your configured LLM — GPT-4o, Claude, DeepSeek, Llama, Gemini, Grok, Mistral, Qwen, or any of 2,050+ available model combinations. Your LLM executes with full context, not guesswork.
Core Components
State Encoder
Converts raw observations into learned representations that capture task-relevant semantics. The foundation everything else is built on — if this is wrong, nothing downstream can fix it.
World Model
Predicts what the environment will look like after a given action — entirely in representation space. This is what makes planning ahead possible without executing anything.
Planning Layer
Combines the world model with a goal-conditioned policy to select and sequence actions. Jointly optimized so prediction and action selection reinforce each other.
Inference Modes
Four modes. One for every decision pattern.Predict State
Simulate what the environment looks like after an action — without executing it. Ask the world model “if I click here, what does the page look like?” and get an answer before committing to anything.
Predict Action
Direct decision making. Given the current state and goal, the planning layer returns the single best next action. No full plan needed — just the right move, right now.
Plan (Multi-Step)
Generate a complete verified action sequence from current state to goal. Returns actions, predicted intermediate states, and a confidence score. Resolves in under 15ms on the small model.
Predict & Verify
The planner proposes. The world model confirms. Only actions where the predicted outcome clears your confidence threshold are executed — with automatic fallback to multi-step planning when they don’t.
When to use Predict State
When to use Predict State
Use this when you want to simulate before committing. Verify safety, expected outcome, or downstream state before an action is applied to the real environment. Ideal for high-cost or irreversible operations.
When to use Predict Action
When to use Predict Action
Use this for direct decision making on known patterns. The planning layer returns its best single action given state and goal — low overhead, fast response. Best when a full plan would be overkill.
When to use Plan (Multi-Step)
When to use Plan (Multi-Step)
Use this when you need a full sequence. The planner runs a forward search through the world model, iteratively selecting actions and simulating outcomes until predicted state converges on the goal or the step budget is exhausted. Returns the ranked sequence, intermediate predicted states, and a confidence score.
When to use Predict & Verify
When to use Predict & Verify
Use this for high-stakes tasks where you need verification before committing. The planner selects a candidate action, the world model simulates the outcome, and the similarity to the goal is scored. If confidence clears the threshold, the action executes. If not, Stratus falls back to a short multi-step plan automatically.
Model Sizes
small
Fast prototyping and low-latency tasks. Under 15ms plan resolution. Pairs well with GPT-4o Mini for high-frequency agent loops.
base
Recommended. The production default — balanced accuracy, latency, and cost. Well-tested across real-world agent workloads.
large
Higher accuracy for complex multi-step tasks. Reach for this when base isn’t enough — not as a default.
xl / huge
Long-horizon planning and demanding environments.
xl for production scale. huge for research and evaluation.API Integration
Stratus wraps planning in an OpenAI-compatible format. Drop it in as a model name — no SDK changes required.Request arrives in standard format
An incoming request arrives in the standard OpenAI chat format. No new protocol. No new SDK.
Model name is parsed
Stratus parses the model name — pattern:
stratus-x1ac-{size}-{llm} — to extract the planning model size and the target downstream LLM.Planning pipeline runs
State is encoded, the world model simulates, the planner sequences, and a verified plan is produced — all before a single token is forwarded.
Execution prompt is enriched
The verified plan is injected into a structured execution prompt, giving your LLM full planning context rather than asking it to reason from scratch.
Pairing the
small planning model with GPT-4o Mini produces an extremely low-latency agent loop suitable for high-frequency tasks. Pairing base with Claude produces a well-balanced production configuration. Mix and match.Why This Architecture Works
Verification before execution
Every action is checked against a predicted outcome before it runs. Not heuristics — a learned model of what actually happens next.
Confidence you can act on
Every plan comes with a confidence score. Route low-confidence plans to fallback logic, escalation, or human review. Build reliability into the loop.
Representation space reasoning
Planning happens in compressed semantic space — not over raw tokens. Faster, more coherent, and less sensitive to surface-level noise in observations.
Fewer retries, more coherence
Multi-step tasks stay on track because each action is planned against the predicted future — not just the present state. Coherence compounds over a task horizon.
Next Steps
Use Cases
See Stratus in action across web navigation, multi-hop reasoning, and task automation.
Quickstart
Build your first Stratus-powered agent in under 10 minutes.
API Reference
Complete endpoint docs, parameters, and response formats.