chini-train-heldout-0001-dp2-workflow

Coffee-Shop Order Pipeline At Morning Rush

easy workflow problem: coffee-shop order pipeline at morning rush

Source: chini-train synth generator v0.1

Prompt

Design a system for: coffee-shop order pipeline at morning rush (domain: ops / physical workflow).

Tier DP2 (easy). 4-6 nodes, single mild stress scenario, moderate criteria.

Constraints:
- At most 8 components on the canvas.
- Monthly cost ceiling: $552 USD. Required behaviors: queue, circuitbreaker.

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: 8
Required behaviors: queue, circuitbreaker
Monthly budget: $552

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.

Pass criteria (overall)

Min stability score: 67
Max drop rate: 13.6%
Min delivery rate: 80.0%
Max errors: 10

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-0001-dp2-workflow \
  --provider openrouter --model google/gemini-2.0-flash-001 \
  --as alice

Or inspect the prompt first:

chini-bench prompt chini-train-heldout-0001-dp2-workflow

Providers: openai · anthropic · google · openrouter · ollama

Leaderboard

Rank	Submitter	Model	Score	Stability	Delivery	Design	Pass
#1	chini-train-08	base_7b single-shot	84	64.0	100.0	75.0	✗
#2	chini-train-08	fmt_a_7b single-shot	84	68.0	74.0	100.0	✗
#3	chini-train-08	base_7b single-shot	84	64.0	100.0	75.0	✗
#4	chini-train-08	fmt_a_7b single-shot	84	68.0	74.0	100.0	✗
#5	chini-train-08	base_7b single-shot	84	64.0	100.0	75.0	✗
#6	chini-train-08	fmt_a_7b single-shot	84	68.0	74.0	100.0	✗
#7	chini-train-08	fmt_a_v5_mixed_7b single-shot	84	68.0	74.0	100.0	✗
#8	chini-train-08	base single-shot	84	64.0	100.0	75.0	✗
#9	chini-train-08	fmt_a_v5 single-shot	84	68.0	74.0	100.0	✗
#10	chini-train-08	rl_v07_pilot_a10b_k8_s1 single-shot	84	68.0	74.0	100.0	✗
#11	chini-train-08	rl_v07_pilot_a10b_k8_s4 single-shot	84	68.0	74.0	100.0	✗
#12	chini-train-08	rl_v07_pilot_a10b_k8_s5 single-shot	84	68.0	74.0	100.0	✗
#13	chini-train-08	rl_v07_pilot_a10b_k8_s7 single-shot	84	68.0	74.0	100.0	✗
#14	chini-train-08	fmtA single-shot	84	68.0	74.0	100.0	✗
#15	chini-train-08	rl_v07_full single-shot	84	68.0	74.0	100.0	✗
#16	chini-train-08	rl_v07_pilot_a10b_k8_s6 single-shot	82	63.0	71.0	100.0	✗
#17	chini-train-08	fmt_a_v4_opus_7b single-shot	81	71.0	78.0	75.0	✗
#18	chini-train-08	fmt_a_v4_opus_7b single-shot	81	71.0	78.0	75.0	✗
#19	chini-train-08	rl_v06_run2 single-shot	81	63.0	68.0	100.0	✗
#20	chini-train-08	rl_v07_pilot_a10b_k8_s0 single-shot	81	71.0	78.0	75.0	✗
#21	chini-train-08	base single-shot	79	61.0	90.0	75.0	✗
#22	chini-train-08	fmt_a_v2 single-shot	79	59.0	62.0	100.0	✗
#23	chini-train-08	base_3b single-shot	79	61.0	90.0	75.0	✗
#24	chini-train-08	fmt_a_3b single-shot	79	59.0	62.0	100.0	✗
#25	chini-train-08	base_3b single-shot	79	61.0	90.0	75.0	✗
#26	chini-train-08	fmt_a_3b single-shot	79	59.0	62.0	100.0	✗
#27	chini-train-08	base single-shot	79	61.0	90.0	75.0	✗
#28	chini-train-08	rl_v07_pilot_a10b_k8_s2 single-shot	78	58.0	61.0	100.0	✗
#29	chini-train-08	rl_v07_pilot_a10b_k8_s3 single-shot	78	58.0	61.0	100.0	✗
#30	chini-train-08	fmt_a single-shot	75	61.0	64.0	75.0	✗

Per-scenario breakdown of the top run

Scenario	Health	Drop rate	Delivered	Pass
baseline	78.0	4.9%	26	✓
adversarial-1	49.0	48.1%	58	✗

How is this scored? →