← back to Model Arena
auto-data-snapshot: 2026-09-21T16:51:55 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
95890732645825c258f38092b58478f2b1f8f74f · 2026-09-21 16:54:55 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit 95890732645825c258f38092b58478f2b1f8f74f
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Mon Sep 21 16:54:55 2026 -0700
auto-data-snapshot: 2026-09-21T16:51:55 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 700 ++++++++++++++++++++++++++++++++++++++++++++-
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 10 +
3 files changed, 700 insertions(+), 12 deletions(-)
diff --git a/data/challenges.json b/data/challenges.json
index 2e53e63..dcab7ad 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -88430,7 +88430,7 @@
"started_at": "2026-09-21T22:54:01.170Z",
"finished_at": null,
"queued_at": "2026-09-21T22:54:00.942Z",
- "resumes": 1
+ "resumes": 2
},
{
"model": "exo__minimax-m2-7-4bit",
@@ -88463,6 +88463,542 @@
"designTools": false,
"created_at": "2026-09-21T23:02:36.498Z",
"winner": null,
+ "runs": [
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 172,
+ "cost": 0,
+ "started_at": "2026-09-21T23:02:36.536Z",
+ "finished_at": "2026-09-21T23:05:28.444Z",
+ "queued_at": "2026-09-21T23:02:36.512Z",
+ "bytes": 10964,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:05:28.458Z",
+ "finished_at": "2026-09-21T23:05:28.619Z",
+ "queued_at": "2026-09-21T23:02:36.524Z"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-21T23:05:39.670Z"
+ },
+ {
+ "id": "17f6791645f9",
+ "title": "History Competition #36 — Spotlit Relic Viewer",
+ "prompt": "Build this as one complete, self-contained HTML file: A museum artifact viewer where an INVENTED relic rotates as a chunky CSS-3D object under a movable spotlight; click facets to reveal provenance and era context, with next/prev and a 'restore vs weathered' toggle. 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": "History",
+ "designTools": false,
+ "created_at": "2026-09-21T23:02:37.462Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 228,
+ "cost": 0,
+ "started_at": "2026-09-21T23:05:28.635Z",
+ "finished_at": "2026-09-21T23:09:16.453Z",
+ "queued_at": "2026-09-21T23:02:37.476Z",
+ "bytes": 13799,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:09:16.467Z",
+ "finished_at": "2026-09-21T23:09:16.704Z",
+ "queued_at": "2026-09-21T23:02:37.488Z"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-21T23:17:43.439Z"
+ },
+ {
+ "id": "2961ba4d6904",
+ "title": "History Competition #36 — Spotlit Relic Viewer",
+ "prompt": "Build this as one complete, self-contained HTML file: A museum artifact viewer where an INVENTED relic rotates as a chunky CSS-3D object under a movable spotlight; click facets to reveal provenance and era context, with next/prev and a 'restore vs weathered' toggle. 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": "History",
+ "designTools": false,
+ "created_at": "2026-09-21T23:10:33.116Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 183,
+ "cost": 0,
+ "started_at": "2026-09-21T23:10:33.255Z",
+ "finished_at": "2026-09-21T23:13:36.330Z",
+ "queued_at": "2026-09-21T23:10:33.133Z",
+ "bytes": 11917,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "error",
+ "error": "timeout after 600s",
+ "seconds": 600,
+ "cost": null,
+ "started_at": "2026-09-21T23:10:33.274Z",
+ "finished_at": "2026-09-21T23:20:33.285Z",
+ "queued_at": "2026-09-21T23:10:33.151Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:13:36.346Z",
+ "finished_at": "2026-09-21T23:13:36.573Z",
+ "queued_at": "2026-09-21T23:10:33.178Z"
+ },
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "error",
+ "error": "timeout after 600s",
+ "seconds": 600,
+ "cost": null,
+ "started_at": "2026-09-21T23:20:33.298Z",
+ "finished_at": "2026-09-21T23:30:33.313Z",
+ "queued_at": "2026-09-21T23:10:33.195Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:13:36.586Z",
+ "finished_at": "2026-09-21T23:13:36.807Z",
+ "queued_at": "2026-09-21T23:10:33.207Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:13:36.821Z",
+ "finished_at": "2026-09-21T23:13:36.975Z",
+ "queued_at": "2026-09-21T23:10:33.230Z"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-21T23:30:33.439Z"
+ },
+ {
+ "id": "91bfdb0020df",
+ "title": "Nature & Space Competition #36 — Shake-the-Snow-Globe Weather",
+ "prompt": "Build this as one complete, self-contained HTML file: A snow-globe weather visualizer you shake (click-drag) to trigger storms over an INVENTED tiny town; a dial cycles sun/rain/storm/snow with smooth transitions and the town reacts (umbrellas, lights). Reduced-motion aware. 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": "Nature & Space",
+ "designTools": false,
+ "created_at": "2026-09-21T23:14:17.726Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 178,
+ "cost": 0,
+ "started_at": "2026-09-21T23:14:17.813Z",
+ "finished_at": "2026-09-21T23:17:15.798Z",
+ "queued_at": "2026-09-21T23:14:17.741Z",
+ "bytes": 11084,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:17:15.827Z",
+ "finished_at": "2026-09-21T23:17:16.018Z",
+ "queued_at": "2026-09-21T23:14:17.756Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:17:16.029Z",
+ "finished_at": "2026-09-21T23:17:16.171Z",
+ "queued_at": "2026-09-21T23:14:17.775Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:17:16.187Z",
+ "finished_at": "2026-09-21T23:17:16.407Z",
+ "queued_at": "2026-09-21T23:14:17.801Z"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-21T23:17:39.887Z"
+ },
+ {
+ "id": "1a0dea3a5b8e",
+ "title": "Comedy Competition #36 — Roast or Toast Campfire",
+ "prompt": "Build this as one complete, self-contained HTML file: A good-natured 'Roast or Toast' campfire where a toggle sends compliments or playful roasts at escalating heat levels shown as literal flames, with shareable text and a 'mercy' button. Keep it kind. 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": "Comedy",
+ "designTools": false,
+ "created_at": "2026-09-21T23:22:22.017Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:22:22.074Z",
+ "finished_at": "2026-09-21T23:22:22.254Z",
+ "queued_at": "2026-09-21T23:22:22.030Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:22:22.266Z",
+ "finished_at": "2026-09-21T23:22:22.418Z",
+ "queued_at": "2026-09-21T23:22:22.041Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 150,
+ "cost": 0,
+ "started_at": "2026-09-21T23:22:22.432Z",
+ "finished_at": "2026-09-21T23:24:52.868Z",
+ "queued_at": "2026-09-21T23:22:22.052Z",
+ "bytes": 9388
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:24:52.880Z",
+ "finished_at": "2026-09-21T23:24:53.090Z",
+ "queued_at": "2026-09-21T23:22:22.063Z"
+ }
+ ]
+ },
+ {
+ "id": "403e6bdabace",
+ "title": "Demoscene Competition #36 — Slime Creature Terrarium",
+ "prompt": "Build this as one complete, self-contained HTML file: Gooey metaballs staged as a terrarium of glowing 'slime creatures' that merge and split; controls for count/threshold/palette, a cursor that repels them, and a snapshot button. 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": "Demoscene",
+ "designTools": false,
+ "created_at": "2026-09-21T23:29:54.225Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 189,
+ "cost": 0,
+ "started_at": "2026-09-21T23:29:54.374Z",
+ "finished_at": "2026-09-21T23:33:03.960Z",
+ "queued_at": "2026-09-21T23:29:54.240Z",
+ "bytes": 13255
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:33:03.972Z",
+ "finished_at": "2026-09-21T23:33:04.143Z",
+ "queued_at": "2026-09-21T23:29:54.252Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:33:04.158Z",
+ "finished_at": "2026-09-21T23:33:04.308Z",
+ "queued_at": "2026-09-21T23:29:54.265Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:33:04.400Z",
+ "finished_at": "2026-09-21T23:33:04.702Z",
+ "queued_at": "2026-09-21T23:29:54.313Z"
+ }
+ ]
+ },
+ {
+ "id": "fcb842231f5e",
+ "title": "Nature & Space Competition #36 — Shake-the-Snow-Globe Weather",
+ "prompt": "Build this as one complete, self-contained HTML file: A snow-globe weather visualizer you shake (click-drag) to trigger storms over an INVENTED tiny town; a dial cycles sun/rain/storm/snow with smooth transitions and the town reacts (umbrellas, lights). Reduced-motion aware. 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": "Nature & Space",
+ "designTools": false,
+ "created_at": "2026-09-21T23:35:34.156Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 215,
+ "cost": 0,
+ "started_at": "2026-09-21T23:35:34.251Z",
+ "finished_at": "2026-09-21T23:39:09.123Z",
+ "queued_at": "2026-09-21T23:35:34.178Z",
+ "bytes": 12253
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:39:09.137Z",
+ "finished_at": "2026-09-21T23:39:09.388Z",
+ "queued_at": "2026-09-21T23:35:34.191Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:39:09.406Z",
+ "finished_at": "2026-09-21T23:39:09.540Z",
+ "queued_at": "2026-09-21T23:35:34.208Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:39:09.553Z",
+ "finished_at": "2026-09-21T23:39:09.707Z",
+ "queued_at": "2026-09-21T23:35:34.237Z"
+ }
+ ]
+ },
+ {
+ "id": "c12a5c7535c9",
+ "title": "Finance Competition #36 — Budget as Jars of Coins",
+ "prompt": "Build this as one complete, self-contained HTML file: A budget planner as a household of jars you pour income into — drag coins between category jars, a 'remaining' jar shows overflow/empty states, validation guards overspend, and a savings-goal jar fills toward a treat. 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": "Finance",
+ "designTools": false,
+ "created_at": "2026-09-21T23:38:06.684Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:39:09.720Z",
+ "finished_at": "2026-09-21T23:39:09.873Z",
+ "queued_at": "2026-09-21T23:38:06.717Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:39:09.898Z",
+ "finished_at": "2026-09-21T23:39:10.059Z",
+ "queued_at": "2026-09-21T23:38:06.729Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 201,
+ "cost": 0,
+ "started_at": "2026-09-21T23:39:10.113Z",
+ "finished_at": "2026-09-21T23:42:30.668Z",
+ "queued_at": "2026-09-21T23:38:06.743Z",
+ "bytes": 12316,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:42:30.681Z",
+ "finished_at": "2026-09-21T23:42:30.932Z",
+ "queued_at": "2026-09-21T23:38:06.764Z"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-21T23:43:15.578Z"
+ },
+ {
+ "id": "5f86343e7db8",
+ "title": "Comedy Competition #36 — Roast or Toast Campfire",
+ "prompt": "Build this as one complete, self-contained HTML file: A good-natured 'Roast or Toast' campfire where a toggle sends compliments or playful roasts at escalating heat levels shown as literal flames, with shareable text and a 'mercy' button. Keep it kind. 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": "Comedy",
+ "designTools": false,
+ "created_at": "2026-09-21T23:44:10.533Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:44:10.596Z",
+ "finished_at": "2026-09-21T23:44:10.787Z",
+ "queued_at": "2026-09-21T23:44:10.545Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:44:10.800Z",
+ "finished_at": "2026-09-21T23:44:10.959Z",
+ "queued_at": "2026-09-21T23:44:10.557Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 133,
+ "cost": 0,
+ "started_at": "2026-09-21T23:44:10.973Z",
+ "finished_at": "2026-09-21T23:46:24.369Z",
+ "queued_at": "2026-09-21T23:44:10.567Z",
+ "bytes": 8488,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:46:24.390Z",
+ "finished_at": "2026-09-21T23:46:24.599Z",
+ "queued_at": "2026-09-21T23:44:10.581Z"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-21T23:49:17.076Z"
+ },
+ {
+ "id": "418fafc484c2",
+ "title": "Design Competition #36 — Pre-Flight Checklist Form Kit",
+ "prompt": "Build this as one complete, self-contained HTML file: An accessible form kit framed as a spaceship pre-flight checklist — validation states light up like console gauges, an inline error summary is the 'do not launch' panel, an optional multi-step flow, and rock-solid keyboard focus. 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": "Design",
+ "designTools": false,
+ "created_at": "2026-09-21T23:47:31.259Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 199,
+ "cost": 0,
+ "started_at": "2026-09-21T23:47:31.334Z",
+ "finished_at": "2026-09-21T23:50:50.549Z",
+ "queued_at": "2026-09-21T23:47:31.273Z",
+ "bytes": 12529
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:50:50.564Z",
+ "finished_at": "2026-09-21T23:50:50.805Z",
+ "queued_at": "2026-09-21T23:47:31.289Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:50:50.848Z",
+ "finished_at": "2026-09-21T23:50:51.001Z",
+ "queued_at": "2026-09-21T23:47:31.307Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T23:50:51.016Z",
+ "finished_at": "2026-09-21T23:50:51.177Z",
+ "queued_at": "2026-09-21T23:47:31.318Z"
+ }
+ ]
+ },
+ {
+ "id": "faf128cfb3c5",
+ "title": "Art Competition #73 — Deep-Sea Fractal Dive",
+ "prompt": "Build this as one complete, self-contained HTML file: A fractal explorer disguised as a deep-sea dive: descend into a Julia set as if sinking through bioluminescent water, a depth gauge ticking, the palette cooling with depth, and a 'creature spotted' note at interesting zoom levels. 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": "Art",
+ "designTools": false,
+ "created_at": "2026-09-21T23:54:16.597Z",
+ "winner": null,
"runs": [
{
"model": "qwen3-14b",
@@ -88470,9 +89006,9 @@
"error": null,
"seconds": null,
"cost": null,
- "started_at": "2026-09-21T23:02:36.536Z",
+ "started_at": "2026-09-21T23:54:16.636Z",
"finished_at": null,
- "queued_at": "2026-09-21T23:02:36.512Z"
+ "queued_at": "2026-09-21T23:54:16.610Z"
},
{
"model": "qwen25-7b",
@@ -88482,17 +89018,17 @@
"cost": null,
"started_at": null,
"finished_at": null,
- "queued_at": "2026-09-21T23:02:36.524Z"
+ "queued_at": "2026-09-21T23:54:16.622Z"
}
]
},
{
- "id": "17f6791645f9",
- "title": "History Competition #36 — Spotlit Relic Viewer",
- "prompt": "Build this as one complete, self-contained HTML file: A museum artifact viewer where an INVENTED relic rotates as a chunky CSS-3D object under a movable spotlight; click facets to reveal provenance and era context, with next/prev and a 'restore vs weathered' toggle. 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": "History",
+ "id": "2f06af4a350c",
+ "title": "Art Competition #73 — Deep-Sea Fractal Dive",
+ "prompt": "Build this as one complete, self-contained HTML file: A fractal explorer disguised as a deep-sea dive: descend into a Julia set as if sinking through bioluminescent water, a depth gauge ticking, the palette cooling with depth, and a 'creature spotted' note at interesting zoom levels. 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": "Art",
"designTools": false,
- "created_at": "2026-09-21T23:02:37.462Z",
+ "created_at": "2026-09-21T23:54:16.655Z",
"winner": null,
"runs": [
{
@@ -88503,7 +89039,7 @@
"cost": null,
"started_at": null,
"finished_at": null,
- "queued_at": "2026-09-21T23:02:37.476Z"
+ "queued_at": "2026-09-21T23:54:16.668Z"
},
{
"model": "qwen25-7b",
@@ -88513,7 +89049,149 @@
"cost": null,
"started_at": null,
"finished_at": null,
- "queued_at": "2026-09-21T23:02:37.488Z"
+ "queued_at": "2026-09-21T23:54:16.681Z"
+ }
+ ]
+ },
+ {
+ "id": "da4c163b454b",
+ "title": "Culture Competition #72 — Lantern Temple by Night",
+ "prompt": "Build this as one complete, self-contained HTML file: A candle-lit scrollytelling tour of an INVENTED heritage temple where scrolling moves a lantern's glow across carvings, revealing a story chapter by chapter, and a constellation completes at the end. 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": "Culture",
+ "designTools": false,
+ "created_at": "2026-09-21T23:54:41.992Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen3-14b",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:42.008Z"
+ },
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-21T23:54:42.105Z",
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:42.021Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:42.038Z"
+ },
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:42.050Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:42.062Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:42.075Z"
+ }
+ ]
+ },
+ {
+ "id": "f8baf0ffaf9c",
+ "title": "News Competition #72 — The City That Floods As You Scroll",
+ "prompt": "Build this as one complete, self-contained HTML file: A data-journalism scrollytelling piece where an INVENTED city floods frame by frame as you drag a rising-sea slider, annotated charts surface, and a final interaction lets the reader 'plant' defenses and watch the chart bend. 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": "News",
+ "designTools": false,
+ "created_at": "2026-09-21T23:54:46.288Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen3-14b",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:46.300Z"
+ },
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:46.312Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:46.324Z"
+ },
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:46.335Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:46.347Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-21T23:54:46.358Z"
}
]
}
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index 431e9b5..cb329e2 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":0,"compCount":{"Art":72,"Culture":71,"News":71,"Political Cartoon":71,"Games":36,"Real Work":36,"Data Viz":36,"Science":36,"Music":36,"Sports":36,"History":36,"Nature & Space":35,"Comedy":35,"Demoscene":35,"Finance":35,"Design":35},"catIdx":11,"perCat":{"Art":3,"Culture":3,"News":3,"Political Cartoon":3,"Games":3,"Real Work":3,"Data Viz":3,"Science":3,"Music":3,"Sports":3,"History":3,"Nature & Space":2,"Comedy":2,"Demoscene":2,"Finance":2,"Design":2}}
\ No newline at end of file
+{"promptIdx":3,"modelOffset":0,"compCount":{"Art":73,"Culture":72,"News":72,"Political Cartoon":71,"Games":36,"Real Work":36,"Data Viz":36,"Science":36,"Music":36,"Sports":36,"History":36,"Nature & Space":36,"Comedy":36,"Demoscene":36,"Finance":36,"Design":36},"catIdx":19,"perCat":{"Art":4,"Culture":4,"News":4,"Political Cartoon":3,"Games":3,"Real Work":3,"Data Viz":3,"Science":3,"Music":3,"Sports":3,"History":3,"Nature & Space":3,"Comedy":3,"Demoscene":3,"Finance":3,"Design":3}}
\ No newline at end of file
diff --git a/yolo/daily-log.jsonl b/yolo/daily-log.jsonl
index ee49950..dc611bf 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -765,3 +765,13 @@
{"ts":"2026-09-21T22:38:43.980Z","id":"413bc0b2523a","promptId":"pomodoro","category":"Real Work","title":"Real Work Competition #36 — Lighthouse Focus Timer","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-21T22:43:52.309Z","id":"d87fb4ee02f4","promptId":"scatter","category":"Data Viz","title":"Data Viz Competition #36 — Planet Dating Show","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-21T22:49:00.699Z","id":"4968a72c2bf2","promptId":"cell","category":"Science","title":"Science Competition #36 — Cell as a Factory-City","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T23:05:32.617Z","id":"0ddb58a78233","promptId":"statcard","category":"Sports","title":"Sports Competition #36 — Holographic Stat Cards","models":2,"done":1,"errored":1,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T23:09:17.711Z","id":"17f6791645f9","promptId":"artifact","category":"History","title":"History Competition #36 — Spotlit Relic Viewer","models":2,"done":1,"errored":1,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T23:17:22.002Z","id":"91bfdb0020df","promptId":"weather","category":"Nature & Space","title":"Nature & Space Competition #36 — Shake-the-Snow-Globe Weather","models":4,"done":1,"errored":3,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T23:24:54.167Z","id":"1a0dea3a5b8e","promptId":"roast","category":"Comedy","title":"Comedy Competition #36 — Roast or Toast Campfire","models":4,"done":1,"errored":3,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T23:30:34.141Z","id":"2961ba4d6904","promptId":"artifact","category":"History","title":"History Competition #36 — Spotlit Relic Viewer","models":6,"done":1,"errored":5,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T23:33:06.630Z","id":"403e6bdabace","promptId":"metaballs","category":"Demoscene","title":"Demoscene Competition #36 — Slime Creature Terrarium","models":4,"done":1,"errored":3,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T23:39:10.523Z","id":"fcb842231f5e","promptId":"weather","category":"Nature & Space","title":"Nature & Space Competition #36 — Shake-the-Snow-Globe Weather","models":4,"done":1,"errored":3,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T23:42:31.241Z","id":"c12a5c7535c9","promptId":"budget","category":"Finance","title":"Finance Competition #36 — Budget as Jars of Coins","models":4,"done":1,"errored":3,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T23:46:26.714Z","id":"5f86343e7db8","promptId":"roast","category":"Comedy","title":"Comedy Competition #36 — Roast or Toast Campfire","models":4,"done":1,"errored":3,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T23:50:51.473Z","id":"418fafc484c2","promptId":"formkit","category":"Design","title":"Design Competition #36 — Pre-Flight Checklist Form Kit","models":4,"done":1,"errored":3,"settled":true,"aiPick":null,"loop":true}
← 980c919 auto-data-snapshot: 2026-09-21T16:01:25 (2 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-21T17:27:28 (3 data files) — dat a7f303c →