Trio-Spark v1.0
Trio-Spark is MachineFi's first Situated World Model: a fast decision model for agents acting inside a specific environment. Give it the current state and 2–8 allowed moves. One pass returns the selected move and a probability for every option.
Trio-Spark is available through a hosted API. Model weights and training code are not distributed from this repository.
Try the playground · API docs · Demos and clients · Launch post
What it does
environment signals → structured state → Trio-Spark → next move → executor
↑ │
└──── fresh feedback ─────┘
Spark handles the bounded judgment step. Your application supplies the state, legal actions, executor, and safety controls. This pattern works across software agents, games, robots, and operational systems.
API
Create an API key in the Trio-Spark console.
curl https://platform.machinefi.com/api/spark/v1/decisions \
-H "Authorization: Bearer $TRIO_SPARK_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $(uuidgen)" \
-d '{
"model": "trio-spark-preview",
"task": "Keep the machine safe while completing the operation.",
"state": {"temperature_c": 84, "load_pct": 91, "vibration": "rising"},
"choices": [
{"id": "continue", "description": "Continue at the current speed"},
{"id": "slow", "description": "Reduce speed and keep observing"},
{"id": "stop", "description": "Stop the machine now"}
]
}'
The response contains choice_id, the complete probability distribution, confidence, usage, and latency. The public API model identifier remains trio-spark-preview; the product release is Trio-Spark v1.0.
Measured examples
The public demo repository includes replayable evidence from production API runs.
| Environment | Result |
|---|---|
| Autonomous drone | 65 s, 25 calls, 75.7 m flown, 0 collisions, course completed |