chini-train-heldout-0008-dp3-civic

Dmv Appointment System At Month-End

moderate civic problem: DMV appointment system at month-end

Source: chini-train synth generator v0.1

Prompt

Design a system for: DMV appointment system at month-end (domain: civic / public service).

Tier DP3 (moderate). 6-9 nodes, two stress scenarios (spike + one of cascade/outage), realistic criteria. Sweet spot.

Constraints:
- At most 10 components on the canvas.
- Monthly cost ceiling: $459 USD. Required behaviors: queue, circuitbreaker, retry.

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: 10
Required behaviors: queue, circuitbreaker, retry
Monthly budget: $459

Stress scenarios

Baseline traffic

baseline

Steady ambient load with no failures.

Adversarial burst

adversarial

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

Cascading failure

cascade

An initial fault propagates through dependent components.

Pass criteria (overall)

Min stability score: 73
Max drop rate: 11.2%
Min delivery rate: 83.7%
Max errors: 8

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-heldout-0008-dp3-civic \
  --provider openrouter --model google/gemini-2.0-flash-001 \
  --as alice

Or inspect the prompt first:

chini-bench prompt chini-train-heldout-0008-dp3-civic

Providers: openai · anthropic · google · openrouter · ollama

Leaderboard

Rank	Submitter	Model	Score	Stability	Delivery	Design	Pass
#1	chini-train-08	rl_v07_pilot_a10b_k8_s5 single-shot	82	55.0	76.0	100.0	✗
#2	chini-train-08	fmt_a single-shot	76	53.0	52.0	100.0	✗
#3	chini-train-08	fmt_a_v2 single-shot	73	47.0	45.0	100.0	✗
#4	chini-train-08	fmt_a_3b single-shot	73	47.0	45.0	100.0	✗
#5	chini-train-08	fmt_a_7b single-shot	73	48.0	46.0	100.0	✗
#6	chini-train-08	fmt_a_3b single-shot	73	47.0	45.0	100.0	✗
#7	chini-train-08	fmt_a_7b single-shot	73	48.0	46.0	100.0	✗
#8	chini-train-08	fmt_a_7b single-shot	73	48.0	46.0	100.0	✗
#9	chini-train-08	fmt_a_v5_mixed_7b single-shot	73	47.0	45.0	100.0	✗
#10	chini-train-08	fmt_a_v5 single-shot	73	47.0	45.0	100.0	✗
#11	chini-train-08	rl_v06_run2 single-shot	73	48.0	46.0	100.0	✗
#12	chini-train-08	rl_v07_pilot_a10b_k8_s0 single-shot	73	47.0	45.0	100.0	✗
#13	chini-train-08	rl_v07_pilot_a10b_k8_s2 single-shot	73	48.0	46.0	100.0	✗
#14	chini-train-08	rl_v07_pilot_a10b_k8_s3 single-shot	73	48.0	46.0	100.0	✗
#15	chini-train-08	rl_v07_pilot_a10b_k8_s6 single-shot	73	47.0	45.0	100.0	✗
#16	chini-train-08	rl_v07_pilot_a10b_k8_s7 single-shot	73	48.0	46.0	100.0	✗
#17	chini-train-08	fmtA single-shot	73	47.0	45.0	100.0	✗
#18	chini-train-08	rl_v07_full single-shot	73	47.0	45.0	100.0	✗
#19	chini-train-08	base_7b single-shot	71	47.0	63.0	75.0	✗
#20	chini-train-08	base_7b single-shot	71	47.0	63.0	75.0	✗
#21	chini-train-08	base_7b single-shot	71	47.0	63.0	75.0	✗
#22	chini-train-08	base single-shot	71	47.0	63.0	75.0	✗
#23	chini-train-08	rl_v07_pilot_a10b_k8_s1 single-shot	71	37.0	53.0	100.0	✗
#24	chini-train-08	fmt_a_v4_opus_7b single-shot	69	51.0	48.0	75.0	✗
#25	chini-train-08	fmt_a_v4_opus_7b single-shot	69	51.0	48.0	75.0	✗
#26	chini-train-08	rl_v07_pilot_a10b_k8_s4 single-shot	68	32.0	42.0	100.0	✗
#27	chini-train-08	base single-shot	65	42.0	66.0	75.0	✗
#28	chini-train-08	base_3b single-shot	65	42.0	66.0	75.0	✗
#29	chini-train-08	base_3b single-shot	65	42.0	66.0	75.0	✗
#30	chini-train-08	base single-shot	65	42.0	66.0	75.0	✗

Per-scenario breakdown of the top run

Scenario	Health	Drop rate	Delivered	Pass
baseline	81.0	1.7%	29	✓
adversarial-1	49.0	87.1%	79	✗
cascade-2	36.0	53.2%	11	✗

How is this scored? →