← back to Model Arena
auto-data-snapshot: 2026-09-20T07:15:24 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
15777c5ebea521f195a4b9b880989c4cfc197c9b · 2026-09-20 07:17:04 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit 15777c5ebea521f195a4b9b880989c4cfc197c9b
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Sun Sep 20 07:17:04 2026 -0700
auto-data-snapshot: 2026-09-20T07:15:24 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 251 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 2 +
3 files changed, 254 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index 6499c4a..b6300aa 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -65194,5 +65194,256 @@
"queued_at": "2026-09-20T13:42:03.930Z"
}
]
+ },
+ {
+ "id": "0b77e4a79e5c",
+ "title": "Games Competition #20 — Wake the Dragon Breakout",
+ "prompt": "Build this as one complete, self-contained HTML file: Brick-breaker where the bricks are a sleeping dragon's scales — knock them loose and the dragon slowly wakes (eyes open, breathing quickens). The paddle is a shield; a caught power-up briefly lets you paint new bricks. Use realistic invented content and a strong information hierarchy. Every visible control must work and give clear feedback; include relevant empty, error, and completion states where applicable. Make it cohesive from 390px to 1440px, keyboard accessible with visible focus, and respectful of reduced motion. Be genuinely creative and surprising — surprise me with the execution, not just the concept.",
+ "category": "Games",
+ "designTools": false,
+ "created_at": "2026-09-20T13:47:11.996Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__llama-3-1-nemotron-70b-instruct-hf-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:47:12.267Z",
+ "finished_at": "2026-09-20T13:47:12.302Z",
+ "queued_at": "2026-09-20T13:47:12.040Z"
+ },
+ {
+ "model": "exo__deepseek-v4-flash",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:47:12.345Z",
+ "finished_at": "2026-09-20T13:47:12.390Z",
+ "queued_at": "2026-09-20T13:47:12.089Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-thinking-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:47:12.438Z",
+ "finished_at": "2026-09-20T13:47:12.497Z",
+ "queued_at": "2026-09-20T13:47:12.134Z"
+ },
+ {
+ "model": "exo__deepseek-v4-pro",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:47:12.523Z",
+ "finished_at": "2026-09-20T13:47:12.571Z",
+ "queued_at": "2026-09-20T13:47:12.158Z"
+ },
+ {
+ "model": "exo__deepseek-v3-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:47:12.613Z",
+ "finished_at": "2026-09-20T13:47:12.642Z",
+ "queued_at": "2026-09-20T13:47:12.196Z"
+ },
+ {
+ "model": "exo__glm-5-8bit-mxfp8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:47:12.689Z",
+ "finished_at": "2026-09-20T13:47:12.738Z",
+ "queued_at": "2026-09-20T13:47:12.235Z"
+ }
+ ]
+ },
+ {
+ "id": "dbf544d44149",
+ "title": "Real Work Competition #20 — Timesheet Garden",
+ "prompt": "Build this as one complete, self-contained HTML file: A timesheet that grows your week as a garden — each logged hour sprouts a plant in the right plot, overtime wilts, and an 'invoice' view harvests it into a clean printable total with rate math and validation. Use realistic invented content and a strong information hierarchy. Every visible control must work and give clear feedback; include relevant empty, error, and completion states where applicable. Make it cohesive from 390px to 1440px, keyboard accessible with visible focus, and respectful of reduced motion. Be genuinely creative and surprising — surprise me with the execution, not just the concept.",
+ "category": "Real Work",
+ "designTools": false,
+ "created_at": "2026-09-20T13:52:20.312Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-mxfp4",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:52:20.584Z",
+ "finished_at": "2026-09-20T13:52:20.691Z",
+ "queued_at": "2026-09-20T13:52:20.353Z"
+ },
+ {
+ "model": "exo__llama-3-1-nemotron-nano-4b-v1-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:52:20.759Z",
+ "finished_at": "2026-09-20T13:52:20.826Z",
+ "queued_at": "2026-09-20T13:52:20.398Z"
+ },
+ {
+ "model": "exo__kimi-k2-6",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:52:20.922Z",
+ "finished_at": "2026-09-20T13:52:20.983Z",
+ "queued_at": "2026-09-20T13:52:20.442Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:52:21.033Z",
+ "finished_at": "2026-09-20T13:52:21.103Z",
+ "queued_at": "2026-09-20T13:52:20.485Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:52:21.149Z",
+ "finished_at": "2026-09-20T13:52:21.216Z",
+ "queued_at": "2026-09-20T13:52:20.513Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 147,
+ "cost": 0,
+ "started_at": "2026-09-20T13:52:20.627Z",
+ "finished_at": "2026-09-20T13:54:47.800Z",
+ "queued_at": "2026-09-20T13:52:20.544Z",
+ "bytes": 9089,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-20T13:54:49.767Z"
+ },
+ {
+ "id": "66bbf490ea69",
+ "title": "Data Viz Competition #20 — Bean-to-Cup Sankey",
+ "prompt": "Build this as one complete, self-contained HTML file: An SVG sankey where an INVENTED coffee empire's beans flow from farm to cup — hover any stream to trace it, click to reroute supply and watch downstream totals update, with an accessible table fallback. Use realistic invented content and a strong information hierarchy. Every visible control must work and give clear feedback; include relevant empty, error, and completion states where applicable. Make it cohesive from 390px to 1440px, keyboard accessible with visible focus, and respectful of reduced motion. Be genuinely creative and surprising — surprise me with the execution, not just the concept.",
+ "category": "Data Viz",
+ "designTools": false,
+ "created_at": "2026-09-20T13:59:52.718Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "error",
+ "error": "timeout after 600s",
+ "seconds": 600,
+ "cost": null,
+ "started_at": "2026-09-20T13:59:52.912Z",
+ "finished_at": "2026-09-20T14:09:52.973Z",
+ "queued_at": "2026-09-20T13:59:52.757Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:59:52.933Z",
+ "finished_at": "2026-09-20T13:59:52.977Z",
+ "queued_at": "2026-09-20T13:59:52.782Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:59:52.953Z",
+ "finished_at": "2026-09-20T13:59:53.138Z",
+ "queued_at": "2026-09-20T13:59:52.820Z"
+ },
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-20T14:09:52.995Z",
+ "finished_at": null,
+ "queued_at": "2026-09-20T13:59:52.842Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:59:52.998Z",
+ "finished_at": "2026-09-20T13:59:53.024Z",
+ "queued_at": "2026-09-20T13:59:52.873Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T13:59:53.246Z",
+ "finished_at": "2026-09-20T13:59:53.531Z",
+ "queued_at": "2026-09-20T13:59:52.893Z"
+ }
+ ]
+ },
+ {
+ "id": "5b861ca6803f",
+ "title": "Daily · Creative Tools: Moodboard Composer",
+ "prompt": "Build this as one complete, self-contained HTML file: A drag-and-drop moodboard with editable swatches, text notes, layout presets, and an export-ready composition. Build for a defined audience and primary journey, using realistic content and a strong information hierarchy. Every visible control must work and return clear feedback; include relevant empty, error, and completion states. Make the result cohesive from 390px to 1440px, keyboard accessible with visible focus, and respectful of reduced motion. Use a high-contrast technical visual direction with excellent information density. Success means a first-time user understands the product, completes the core task, and can verify the outcome without instructions.",
+ "category": "Creative Tools",
+ "designTools": false,
+ "created_at": "2026-09-20T14:15:00.789Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen3-14b",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-20T14:15:00.879Z",
+ "finished_at": null,
+ "queued_at": "2026-09-20T14:15:00.823Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-20T14:15:00.849Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index dd4d480..86e7e0f 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":118,"compCount":{"Art":40,"Culture":39,"News":39,"Political Cartoon":39,"Games":19,"Real Work":19,"Data Viz":19,"Science":19,"Music":19,"Sports":19,"History":19,"Nature & Space":19,"Comedy":19,"Demoscene":19,"Finance":19,"Design":19},"catIdx":4,"perCat":{"Art":3,"Culture":3,"News":3,"Political Cartoon":3,"Games":1,"Real Work":1,"Data Viz":1,"Science":1,"Music":1,"Sports":1,"History":1,"Nature & Space":1,"Comedy":1,"Demoscene":1,"Finance":1,"Design":1}}
\ No newline at end of file
+{"promptIdx":3,"modelOffset":7,"compCount":{"Art":40,"Culture":39,"News":39,"Political Cartoon":39,"Games":20,"Real Work":20,"Data Viz":20,"Science":19,"Music":19,"Sports":19,"History":19,"Nature & Space":19,"Comedy":19,"Demoscene":19,"Finance":19,"Design":19},"catIdx":7,"perCat":{"Art":3,"Culture":3,"News":3,"Political Cartoon":3,"Games":2,"Real Work":2,"Data Viz":2,"Science":1,"Music":1,"Sports":1,"History":1,"Nature & Space":1,"Comedy":1,"Demoscene":1,"Finance":1,"Design":1}}
\ No newline at end of file
diff --git a/yolo/daily-log.jsonl b/yolo/daily-log.jsonl
index 0f2a50b..24ae521 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -439,3 +439,5 @@
{"ts":"2026-09-20T13:31:55.408Z","id":"5fea87383f41","promptId":"languages","category":"Culture","title":"Culture Competition #39 — The Last Speaker","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-20T13:37:03.692Z","id":"c030f1937ce0","promptId":"news-ticker","category":"News","title":"News Competition #39 — Absurd 24-Hour News","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-20T13:42:11.953Z","id":"c409f538fd93","promptId":"metaphor","category":"Political Cartoon","title":"Political Cartoon Competition #39 — Ship of State by Committee","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T13:47:20.272Z","id":"0b77e4a79e5c","promptId":"breakout","category":"Games","title":"Games Competition #20 — Wake the Dragon Breakout","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T13:54:52.661Z","id":"dbf544d44149","promptId":"timesheet","category":"Real Work","title":"Real Work Competition #20 — Timesheet Garden","models":6,"done":1,"errored":5,"settled":true,"aiPick":null,"loop":true}
← 5f64012 auto-data-snapshot: 2026-09-20T06:42:09 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-20T07:48:13 (3 data files) — dat 59d55f7 →