A silver spacecraft banks past asteroids over a glowing mint grid.

A SMALL MODEL. EVERY MOVE MATTERS.

Human instinct.
Machine decisions.

One grid. Falling hazards. A choice that matters.
Watch Strands Decider 2B choose its next move.
Then explore how we built it.

80-second recorded walkthrough

THE AI ARCADE / BUILT BY ELYAS SYEDExplore the experiment

MEET THE MODEL

What is
Strands Decider 2B?

Strands Decider 2B is a small, open-source decision model from AWS, optimized for fast experimentation, local development, and innovation. Give it context, a question, and a set of options. It returns a choice, yes/no answer, or score.

In Grid Dodger, it chooses left, stay, or right. The game executes the move, and Amazon Bedrock writes the end-of-run recap.

01 / THE EXPERIMENT

Same grid.
A different kind of pilot.

Most AI demos ask you to read an answer.
This one lets you watch a decision play out.

01

Human

Trust your instincts.

You steer. Read the falling hazards, choose a lane, and see how far you can go.

YOU FLY
02

Decider

Hand over the controls.

The model chooses left, stay, or right. A registered movement tool executes the choice.

AI FLIES3 OPTIONS → 1 TOOL
03

Shadow

Keep the final say.

Decider suggests the next move before you act. Follow its lead—or follow your gut.

AI GUIDES. YOU DECIDE.

Five lanes. One goal: 500 points.
Difficulty starts rising at 100. Every move advances the game.

Watch the demo

02 / INSIDE THE AGENT

Small model.
A meaningful decision.

Strands Decider 2B is a small decision model.
Give it a question and a fixed set of options.
It chooses; your application acts.

THE NEXT MOVE
Player ship

LEFT · STAY · RIGHT

THE DECISION RECIPE

01 / GROUNDED CONTEXT

Read the board.

The engine describes the board and calculates each move’s immediate outcome. Decider receives explicit facts, not an image to guess from.

move-selection.recipe
// a bounded question, three registered tools
question: "Which move is safest?"
options:  ["left", "stay", "right"]
selected: "right"
tool:     move_right()

Decider selects. Deterministic code dispatches the tool.
Amazon Bedrock narrates the finished run.

THE STACKStrands Decider 2B+Amazon Bedrock+Browser game engine

03 / WHAT WE LEARNED

Better context.
Better next moves.

The useful finding wasn’t a bigger model.
It was a clearer question, grounded in the game.

17 → 0

Avoidable collisions

Original vs. grounded prompt,
on 31 avoidable snapshot cases.

422ms

Median decision request

Grounded prompt, 40 snapshots.
Includes HTTP and network time.

51/60

Mean episode length

Grounded prompt, five seeded runs.
Episodes capped at 60 moves.

Measured, side by side.

40 fixed snapshots · 5 seeded runs per variant · 60-move cap

Recorded October 17, 2026. The same snapshot cases and spawn seeds were used for each variant, at a 0.40 fallback threshold.
VariantAvoidable
collisions ↓
Mean survival
moves ↑
Reached
60 moves ↑
Median
latency ↓
p95
latency ↓
Errors
HeuristicLocal rule-based baseline0 / 3144.82 / 50.02 ms0.10 ms0
Original promptDecider · reconstructed earlier question17 / 3124.20 / 5354.66 ms384.74 ms0
Grounded promptDecider · explicit move outcomes0 / 3151.04 / 5422.06 ms519.09 ms0

Lower collisions and latency are better; higher survival is better. p95 is the time at or below which 95% of snapshot requests finished. Runs reaching 60 moves were stopped at the test limit.

What it means: Explicit move outcomes helped Decider avoid immediate mistakes and survive longer in this small test, with a higher median request time. The rule-based baseline also avoided every avoidable snapshot collision and was much faster.

Read the benchmark and its limits (opens in a new tab)

04 / WATCH THE DECISIONS PLAY OUT

A small model.
In the pilot’s seat.

See the board. Watch the choice.
Follow the move that happens next.

GRID DODGER / DECIDER MODERECORDED GAMEPLAY
Previously captured real Decider gameplay at 44 points, showing its chosen move, probabilities, and decision time.
CAPTURED GAMEPLAY · STILL IMAGE

Watch the recorded run.

This is a real Decider run.
See Human, Decider, and Shadow modes in the video.

Recorded Human, Decider, and Shadow gameplay with a real Amazon Bedrock recap.

Build it yourself

WHAT TO WATCH FOR

01The board gives the model grounded move outcomes.

02The choice selects left, stay, or right.

03The tool turns that choice into a move.

05 / BUILD IT YOURSELF

Take the idea.
Make your next move.

Run Grid Dodger with real Strands Decider 2B on your own computer, or deploy it in your AWS account. Local movement decisions need no AWS hosting; speed depends on your hardware. Bedrock recaps are optional and use paid AWS model calls.