chini-train-train-0373-dp5-civic

Food-Bank Distribution Under Snap Delay

brutal civic problem: food-bank distribution under SNAP delay

Source: chini-train synth generator v0.1

Prompt

Design a system for: food-bank distribution under SNAP delay (domain: civic / public service).

Tier DP5 (brutal). 9-12 nodes, four scenarios with high intensity, brutal criteria. Failing examples.

Constraints:
- At most 12 components on the canvas.
- Monthly cost ceiling: $273 USD. Required behaviors: queue, circuitbreaker, retry, ratelimit.

Return a Chinilla CanvasState that handles the listed scenarios. Include trigger components for each entry point and at least one terminal storage / sink so the simulator can score delivery.

Constraints

Max components: 12
Required behaviors: queue, circuitbreaker, retry, ratelimit
Monthly budget: $273

Stress scenarios

Baseline traffic

baseline

Steady ambient load with no failures.

Traffic spike

spike

Traffic suddenly multiplies. The hot path must hold.

Adversarial burst

adversarial

Hostile packets injected on top of clean traffic. Defenses must block them without dropping good requests.

Dependency outage

outage

A downstream component is disabled. System must degrade gracefully.

Cascading failure

cascade

An initial fault propagates through dependent components.

Pass criteria (overall)

Min stability score: 86
Max drop rate: 6.4%
Min delivery rate: 91.2%
Max errors: 4

Submit your run

Submissions go through the chini-bench CLI. It calls your model with your key, scores the result locally, and posts to the leaderboard. Nothing leaves your machine except the canvas it produces.

End-to-end:

pip install git+https://github.com/collapseindex/chini-bench-cli.git
export OPENROUTER_API_KEY=...

chini-bench run chini-train-train-0373-dp5-civic \
  --provider openrouter --model google/gemini-2.0-flash-001 \
  --as alice

Or inspect the prompt first:

chini-bench prompt chini-train-train-0373-dp5-civic

Providers: openai · anthropic · google · openrouter · ollama

Leaderboard

Rank	Submitter	Model	Score	Stability	Delivery	Design	Pass
#1	rl_v07_full_a10	rl_policy custom single-shot	82	60.0	59.0	100.0	✗
#2	rl_v07_full_a10	rl_policy custom single-shot	81	58.0	58.0	100.0	✗
#3	rl_v07_full_a10	rl_policy custom single-shot	81	58.0	57.0	100.0	✗
#4	rl_v07_full_a10	rl_policy custom single-shot	81	60.0	58.0	100.0	✗
#5	chini-train-03	grok-4.1-fast single-shot	66	29.0	22.0	100.0	✗
#6	chini-train-04	grok-4.1-fast single-shot	66	29.0	22.0	100.0	✗

Per-scenario breakdown of the top run

Scenario	Health	Drop rate	Delivered	Pass
baseline	82.0	1.7%	29	✗
spike-1	57.0	21.0%	1097	✗
adversarial-2	50.0	90.8%	107	✗
outage-3	73.0	2.8%	0	✗
cascade-4	39.0	50.0%	12	✗

How is this scored? →