chini-train-train-0029-dp2-adversarial

Rate-Limited Search Under Enumeration Attack

easy adversarial problem: rate-limited search under enumeration attack

Source: chini-train synth generator v0.1

Prompt

Design a system for: rate-limited search under enumeration attack (domain: adversarial).

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

Constraints:
- At most 8 components on the canvas.
- Monthly cost ceiling: $580 USD. Required behaviors: ratelimit, 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: ratelimit, queue, circuitbreaker
Monthly budget: $580

Stress scenarios

Baseline traffic

baseline

Steady ambient load with no failures.

Traffic spike

spike

Traffic suddenly multiplies. The hot path must hold.

Pass criteria (overall)

Min stability score: 58
Max drop rate: 14.0%
Min delivery rate: 75.8%
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-train-0029-dp2-adversarial \
  --provider openrouter --model google/gemini-2.0-flash-001 \
  --as alice

Or inspect the prompt first:

chini-bench prompt chini-train-train-0029-dp2-adversarial

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	88	76.0	87.0	100.0	✓
#2	rl_v07_pilot_a10b	rl_policy custom single-shot	87	73.0	84.0	100.0	✗
#3	chini-train-03	opus-4.7 single-shot	85	71.0	80.0	100.0	✗
#4	chini-train-03	opus-4.7 single-shot	85	71.0	80.0	100.0	✗
#5	rl_v07_full_a10	rl_policy custom single-shot	82	66.0	75.0	100.0	✗
#6	rl_v07_full_a10	rl_policy custom single-shot	82	66.0	74.0	100.0	✗
#7	rl_v07_pilot_a10b	rl_policy custom single-shot	81	65.0	73.0	100.0	✗
#8	rl_v07_pilot_a10b	rl_policy custom single-shot	81	65.0	73.0	100.0	✗
#9	chini-train-03	grok-4.1-fast single-shot	80	63.0	72.0	100.0	✗
#10	rl_v07_pilot_a10b	rl_policy custom single-shot	80	63.0	70.0	100.0	✗
#11	rl_v07_full_a10	rl_policy custom single-shot	76	57.0	63.0	100.0	✗

Per-scenario breakdown of the top run

Scenario	Health	Drop rate	Delivered	Pass
baseline	84.0	1.7%	29	✓
spike-1	67.0	13.2%	614	✓

How is this scored? →