chini-train-heldout-0000-dp1-infra

Url Shortener

trivial infra problem: URL shortener

Source: chini-train synth generator v0.1

Prompt

Design a system for: URL shortener (domain: SWE backend / infra).

Tier DP1 (trivial). 3-5 nodes, baseline scenario only, lax criteria. Sanity floor.

Constraints:
- At most 6 components on the canvas.
- Monthly cost ceiling: $676 USD. Required behaviors: storage, queue.

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: 6
Required behaviors: storage, queue
Monthly budget: $676

Stress scenarios

Baseline traffic

baseline

Steady ambient load with no failures.

Pass criteria (overall)

Min stability score: 58
Max drop rate: 16.8%
Min delivery rate: 75.0%
Max errors: 12

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-0000-dp1-infra \
  --provider openrouter --model google/gemini-2.0-flash-001 \
  --as alice

Or inspect the prompt first:

chini-bench prompt chini-train-heldout-0000-dp1-infra

Providers: openai · anthropic · google · openrouter · ollama

Leaderboard

Rank	Submitter	Model	Score	Stability	Delivery	Design	Pass
#1	chini-train-08	fmt_a_v2 single-shot	96	87.0	100.0	100.0	✓
#2	chini-train-08	fmt_a_3b single-shot	96	87.0	100.0	100.0	✓
#3	chini-train-08	base_7b single-shot	96	87.0	100.0	100.0	✓
#4	chini-train-08	fmt_a_7b single-shot	96	87.0	100.0	100.0	✓
#5	chini-train-08	fmt_a_3b single-shot	96	87.0	100.0	100.0	✓
#6	chini-train-08	base_7b single-shot	96	87.0	100.0	100.0	✓
#7	chini-train-08	fmt_a_7b single-shot	96	87.0	100.0	100.0	✓
#8	chini-train-08	fmt_a_v4_opus_7b single-shot	96	88.0	100.0	100.0	✓
#9	chini-train-08	base_7b single-shot	96	87.0	100.0	100.0	✓
#10	chini-train-08	fmt_a_7b single-shot	96	87.0	100.0	100.0	✓
#11	chini-train-08	fmt_a_v4_opus_7b single-shot	96	88.0	100.0	100.0	✓
#12	chini-train-08	fmt_a_v5_mixed_7b single-shot	96	87.0	100.0	100.0	✓
#13	chini-train-08	base single-shot	96	87.0	100.0	100.0	✓
#14	chini-train-08	fmt_a_v5 single-shot	96	87.0	100.0	100.0	✓
#15	chini-train-08	rl_v06_run2 single-shot	96	87.0	100.0	100.0	✓
#16	chini-train-08	rl_v07_pilot_a10b_k8_s0 single-shot	96	87.0	100.0	100.0	✓
#17	chini-train-08	rl_v07_pilot_a10b_k8_s1 single-shot	96	87.0	100.0	100.0	✓
#18	chini-train-08	rl_v07_pilot_a10b_k8_s2 single-shot	96	87.0	100.0	100.0	✓
#19	chini-train-08	rl_v07_pilot_a10b_k8_s3 single-shot	96	87.0	100.0	100.0	✓
#20	chini-train-08	rl_v07_pilot_a10b_k8_s4 single-shot	96	87.0	100.0	100.0	✓
#21	chini-train-08	rl_v07_pilot_a10b_k8_s5 single-shot	96	87.0	100.0	100.0	✓
#22	chini-train-08	rl_v07_pilot_a10b_k8_s6 single-shot	96	87.0	100.0	100.0	✓
#23	chini-train-08	rl_v07_pilot_a10b_k8_s7 single-shot	96	87.0	100.0	100.0	✓
#24	chini-train-08	fmtA single-shot	96	87.0	100.0	100.0	✓
#25	chini-train-08	rl_v07_full single-shot	96	87.0	100.0	100.0	✓
#26	chini-train-08	fmt_a single-shot	81	87.0	50.0	100.0	✗

Per-scenario breakdown of the top run

Scenario	Health	Drop rate	Delivered	Pass
baseline	87.0	0.0%	15	✓

How is this scored? →