← back to Model Arena
auto-data-snapshot: 2026-09-19T22:17:38 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
b04b8b0940230f07e24ba312d4c8a2c3d16f015d · 2026-09-19 22:19:55 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit b04b8b0940230f07e24ba312d4c8a2c3d16f015d
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Sat Sep 19 22:19:55 2026 -0700
auto-data-snapshot: 2026-09-19T22:17:38 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 362 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 5 +
3 files changed, 368 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index b886c81..f3ffcc8 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -58918,5 +58918,367 @@
"queued_at": "2026-09-20T04:46:19.615Z"
}
]
+ },
+ {
+ "id": "4593df1a89c5",
+ "title": "Art Competition #31 — Playable Mondrian",
+ "prompt": "Build this as one complete, self-contained HTML file: A Mondrian composition you can play like an instrument: clicking each colored rectangle triggers a tone and a ripple, and a 'let it compose itself' mode slowly rearranges the grid to a generative rhythm. 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": "2026-09-20T04:51:27.664Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__qwen3-5-397b-a17b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:51:27.843Z",
+ "finished_at": "2026-09-20T04:51:27.884Z",
+ "queued_at": "2026-09-20T04:51:27.692Z"
+ },
+ {
+ "model": "exo__glm-4-5-air-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:51:27.907Z",
+ "finished_at": "2026-09-20T04:51:27.941Z",
+ "queued_at": "2026-09-20T04:51:27.718Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:51:27.965Z",
+ "finished_at": "2026-09-20T04:51:27.991Z",
+ "queued_at": "2026-09-20T04:51:27.753Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-nvfp4",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:51:28.017Z",
+ "finished_at": "2026-09-20T04:51:28.059Z",
+ "queued_at": "2026-09-20T04:51:27.779Z"
+ },
+ {
+ "model": "exo__qwen3-6-27b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:51:28.086Z",
+ "finished_at": "2026-09-20T04:51:28.128Z",
+ "queued_at": "2026-09-20T04:51:27.807Z"
+ },
+ {
+ "model": "exo__llama-3-3-70b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:51:28.166Z",
+ "finished_at": "2026-09-20T04:51:28.188Z",
+ "queued_at": "2026-09-20T04:51:27.825Z"
+ }
+ ]
+ },
+ {
+ "id": "e675a6a79e20",
+ "title": "Culture Competition #30 — Festivals of Nowhere",
+ "prompt": "Build this as one complete, self-contained HTML file: A world map of INVENTED holidays where hovering a region launches its signature celebration in miniature (lanterns, confetti, fireworks) and a little passport stamps each one you visit. 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": "2026-09-20T04:56:35.908Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__glm-5-1-mxfp4-q8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:56:36.154Z",
+ "finished_at": "2026-09-20T04:56:36.209Z",
+ "queued_at": "2026-09-20T04:56:35.951Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:56:36.227Z",
+ "finished_at": "2026-09-20T04:56:36.248Z",
+ "queued_at": "2026-09-20T04:56:35.992Z"
+ },
+ {
+ "model": "exo__minimax-m2-5-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:56:36.265Z",
+ "finished_at": "2026-09-20T04:56:36.289Z",
+ "queued_at": "2026-09-20T04:56:36.031Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:56:36.309Z",
+ "finished_at": "2026-09-20T04:56:36.335Z",
+ "queued_at": "2026-09-20T04:56:36.074Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:56:36.355Z",
+ "finished_at": "2026-09-20T04:56:36.381Z",
+ "queued_at": "2026-09-20T04:56:36.096Z"
+ },
+ {
+ "model": "exo__kimi-k2-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T04:56:36.398Z",
+ "finished_at": "2026-09-20T04:56:36.418Z",
+ "queued_at": "2026-09-20T04:56:36.114Z"
+ }
+ ]
+ },
+ {
+ "id": "1fbe7c6dec1c",
+ "title": "News Competition #30 — Teletext Econ Briefing",
+ "prompt": "Build this as one complete, self-contained HTML file: A retro-teletext economic briefing for an INVENTED country: blocky CRT panels tick with fake indicators, a crawler scrolls, a big red/green mood dial summarizes the day — with scanlines and a channel-flip. 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": "2026-09-20T05:01:44.198Z",
+ "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": "2026-09-20T05:01:44.459Z",
+ "finished_at": "2026-09-20T05:01:44.497Z",
+ "queued_at": "2026-09-20T05:01:44.243Z"
+ },
+ {
+ "model": "exo__deepseek-v4-flash",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:01:44.539Z",
+ "finished_at": "2026-09-20T05:01:44.562Z",
+ "queued_at": "2026-09-20T05:01:44.276Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-thinking-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:01:44.587Z",
+ "finished_at": "2026-09-20T05:01:44.644Z",
+ "queued_at": "2026-09-20T05:01:44.313Z"
+ },
+ {
+ "model": "exo__deepseek-v4-pro",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:01:44.670Z",
+ "finished_at": "2026-09-20T05:01:44.723Z",
+ "queued_at": "2026-09-20T05:01:44.357Z"
+ },
+ {
+ "model": "exo__deepseek-v3-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:01:44.771Z",
+ "finished_at": "2026-09-20T05:01:44.799Z",
+ "queued_at": "2026-09-20T05:01:44.400Z"
+ },
+ {
+ "model": "exo__glm-5-8bit-mxfp8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:01:44.862Z",
+ "finished_at": "2026-09-20T05:01:44.903Z",
+ "queued_at": "2026-09-20T05:01:44.428Z"
+ }
+ ]
+ },
+ {
+ "id": "7bdc7d0a2ddc",
+ "title": "Political Cartoon Competition #30 — The Form to Request a Form",
+ "prompt": "Build this as one complete, self-contained HTML file: A three-panel SVG satire strip about an INVENTED bureaucratic saga (a form that requires a form to request a form). Click to advance panels; a 'plot twist' button reshuffles the punchline. Good-natured, invented characters. 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": "Political Cartoon",
+ "designTools": false,
+ "created_at": "2026-09-20T05:06:52.496Z",
+ "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": "2026-09-20T05:06:52.736Z",
+ "finished_at": "2026-09-20T05:06:52.813Z",
+ "queued_at": "2026-09-20T05:06:52.537Z"
+ },
+ {
+ "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": "2026-09-20T05:06:52.842Z",
+ "finished_at": "2026-09-20T05:06:52.869Z",
+ "queued_at": "2026-09-20T05:06:52.580Z"
+ },
+ {
+ "model": "exo__kimi-k2-6",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:06:52.886Z",
+ "finished_at": "2026-09-20T05:06:52.919Z",
+ "queued_at": "2026-09-20T05:06:52.624Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:06:52.941Z",
+ "finished_at": "2026-09-20T05:06:52.968Z",
+ "queued_at": "2026-09-20T05:06:52.665Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:06:52.998Z",
+ "finished_at": "2026-09-20T05:06:53.021Z",
+ "queued_at": "2026-09-20T05:06:52.698Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 124,
+ "cost": 0,
+ "started_at": "2026-09-20T05:06:52.771Z",
+ "finished_at": "2026-09-20T05:08:56.825Z",
+ "queued_at": "2026-09-20T05:06:52.716Z",
+ "bytes": 7485,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-20T05:08:58.844Z"
+ },
+ {
+ "id": "62d5208c7aae",
+ "title": "Art Competition #32 — Swarm Typography Poster",
+ "prompt": "Build this as one complete, self-contained HTML file: A kinetic-type poster for an INVENTED avant-garde festival where the headline letters behave like a flocking swarm — scattering from the cursor and re-forming the title — with a slider that morphs the mood from 'whisper' to 'scream'. 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": "2026-09-20T05:14:00.850Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-20T05:14:00.973Z",
+ "finished_at": null,
+ "queued_at": "2026-09-20T05:14:00.870Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:14:00.991Z",
+ "finished_at": "2026-09-20T05:14:01.027Z",
+ "queued_at": "2026-09-20T05:14:00.888Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:14:01.007Z",
+ "finished_at": "2026-09-20T05:14:01.189Z",
+ "queued_at": "2026-09-20T05:14:00.906Z"
+ },
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-20T05:14:00.923Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:14:01.044Z",
+ "finished_at": "2026-09-20T05:14:01.067Z",
+ "queued_at": "2026-09-20T05:14:00.940Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T05:14:01.219Z",
+ "finished_at": "2026-09-20T05:14:01.378Z",
+ "queued_at": "2026-09-20T05:14:00.957Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index 0742141..031bec2 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":106,"compCount":{"Art":30,"Culture":29,"News":29,"Political Cartoon":29,"Games":15,"Real Work":15,"Data Viz":15,"Science":15,"Music":15,"Sports":15,"History":15,"Nature & Space":15,"Comedy":15,"Demoscene":15,"Finance":15,"Design":15},"catIdx":16,"perCat":{"Art":1,"Culture":1,"News":1,"Political Cartoon":1,"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
+{"promptIdx":3,"modelOffset":7,"compCount":{"Art":32,"Culture":30,"News":30,"Political Cartoon":30,"Games":15,"Real Work":15,"Data Viz":15,"Science":15,"Music":15,"Sports":15,"History":15,"Nature & Space":15,"Comedy":15,"Demoscene":15,"Finance":15,"Design":15},"catIdx":1,"perCat":{"Art":3,"Culture":2,"News":2,"Political Cartoon":2,"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 7d898f5..511154b 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -350,3 +350,8 @@
{"ts":"2026-09-20T04:31:02.918Z","id":"71fdb3fa16aa","promptId":"roast","category":"Comedy","title":"Comedy Competition #15 — Roast or Toast Campfire","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-20T04:36:11.139Z","id":"ed9fbd81160c","promptId":"metaballs","category":"Demoscene","title":"Demoscene Competition #15 — Slime Creature Terrarium","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-20T04:41:19.368Z","id":"972ba9308bbc","promptId":"budget","category":"Finance","title":"Finance Competition #15 — Budget as Jars of Coins","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T04:46:27.642Z","id":"90a053682442","promptId":"formkit","category":"Design","title":"Design Competition #15 — Pre-Flight Checklist Form Kit","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T04:51:35.848Z","id":"4593df1a89c5","promptId":"mondrian","category":"Art","title":"Art Competition #31 — Playable Mondrian","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T04:56:44.158Z","id":"e675a6a79e20","promptId":"culture-map","category":"Culture","title":"Culture Competition #30 — Festivals of Nowhere","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T05:01:52.464Z","id":"1fbe7c6dec1c","promptId":"econ-brief","category":"News","title":"News Competition #30 — Teletext Econ Briefing","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T05:09:00.789Z","id":"7bdc7d0a2ddc","promptId":"satire-strip","category":"Political Cartoon","title":"Political Cartoon Competition #30 — The Form to Request a Form","models":6,"done":1,"errored":5,"settled":true,"aiPick":null,"loop":true}
← 7fd2651 auto-data-snapshot: 2026-09-19T21:44:50 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-19T22:51:15 (3 data files) — dat 75692fb →