← back to Model Arena
auto-data-snapshot: 2026-09-19T13:42:07 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
a6656b9319284f2a2f1476dc88ef2e79954ded6a · 2026-09-19 13:43:11 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit a6656b9319284f2a2f1476dc88ef2e79954ded6a
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Sat Sep 19 13:43:11 2026 -0700
auto-data-snapshot: 2026-09-19T13:42:07 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 291 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 3 +
3 files changed, 295 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index 310d79b..c9ed8ef 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -52862,5 +52862,296 @@
"queued_at": "2026-09-19T20:06:34.465Z"
}
]
+ },
+ {
+ "id": "7511f1a0843c",
+ "title": "Nature & Space Competition #11 — Terrarium in a Jar",
+ "prompt": "Build this as one complete, self-contained HTML file: A jar-terrarium ecosystem of INVENTED critters with live population graphs — feed them, introduce a predator, or trigger a drought and watch the food web wobble toward chaos or balance, with a reset. 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": "2026-09-19T20:11:42.546Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__glm-5-1-mxfp4-q8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:11:42.689Z",
+ "finished_at": "2026-09-19T20:11:42.712Z",
+ "queued_at": "2026-09-19T20:11:42.571Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:11:42.729Z",
+ "finished_at": "2026-09-19T20:11:42.754Z",
+ "queued_at": "2026-09-19T20:11:42.595Z"
+ },
+ {
+ "model": "exo__minimax-m2-5-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:11:42.777Z",
+ "finished_at": "2026-09-19T20:11:42.796Z",
+ "queued_at": "2026-09-19T20:11:42.617Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:11:42.818Z",
+ "finished_at": "2026-09-19T20:11:42.886Z",
+ "queued_at": "2026-09-19T20:11:42.639Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:11:42.919Z",
+ "finished_at": "2026-09-19T20:11:42.943Z",
+ "queued_at": "2026-09-19T20:11:42.657Z"
+ },
+ {
+ "model": "exo__kimi-k2-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:11:42.965Z",
+ "finished_at": "2026-09-19T20:11:42.986Z",
+ "queued_at": "2026-09-19T20:11:42.673Z"
+ }
+ ]
+ },
+ {
+ "id": "58121b9214db",
+ "title": "Comedy Competition #11 — The Excuse Slot Machine",
+ "prompt": "Build this as one complete, self-contained HTML file: An 'Excuse Engine' slot machine that spins three reels (who / what / cosmic reason) into gloriously implausible excuses, with a believability meter, a 'too far' lever, favorites, and copy-to-clipboard. 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": "2026-09-19T20:16:50.751Z",
+ "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": "2026-09-19T20:16:50.869Z",
+ "finished_at": "2026-09-19T20:16:50.889Z",
+ "queued_at": "2026-09-19T20:16:50.770Z"
+ },
+ {
+ "model": "exo__deepseek-v4-flash",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:16:50.904Z",
+ "finished_at": "2026-09-19T20:16:50.923Z",
+ "queued_at": "2026-09-19T20:16:50.787Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-thinking-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:16:50.941Z",
+ "finished_at": "2026-09-19T20:16:50.961Z",
+ "queued_at": "2026-09-19T20:16:50.803Z"
+ },
+ {
+ "model": "exo__deepseek-v4-pro",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:16:50.978Z",
+ "finished_at": "2026-09-19T20:16:50.998Z",
+ "queued_at": "2026-09-19T20:16:50.820Z"
+ },
+ {
+ "model": "exo__deepseek-v3-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:16:51.015Z",
+ "finished_at": "2026-09-19T20:16:51.034Z",
+ "queued_at": "2026-09-19T20:16:50.836Z"
+ },
+ {
+ "model": "exo__glm-5-8bit-mxfp8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:16:51.051Z",
+ "finished_at": "2026-09-19T20:16:51.071Z",
+ "queued_at": "2026-09-19T20:16:50.852Z"
+ }
+ ]
+ },
+ {
+ "id": "1d0dbe6f8207",
+ "title": "Demoscene Competition #11 — Wormhole to Anywhere",
+ "prompt": "Build this as one complete, self-contained HTML file: An infinite tunnel that flies through an INVENTED wormhole to shifting destinations (neon city, coral reef, circuit board) selectable on the fly, with texture/palette controls and auto-motion. 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": "2026-09-19T20:21:58.905Z",
+ "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": "2026-09-19T20:21:59.042Z",
+ "finished_at": "2026-09-19T20:21:59.077Z",
+ "queued_at": "2026-09-19T20:21:58.925Z"
+ },
+ {
+ "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": "2026-09-19T20:21:59.094Z",
+ "finished_at": "2026-09-19T20:21:59.114Z",
+ "queued_at": "2026-09-19T20:21:58.946Z"
+ },
+ {
+ "model": "exo__kimi-k2-6",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:21:59.131Z",
+ "finished_at": "2026-09-19T20:21:59.152Z",
+ "queued_at": "2026-09-19T20:21:58.966Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:21:59.171Z",
+ "finished_at": "2026-09-19T20:21:59.191Z",
+ "queued_at": "2026-09-19T20:21:58.983Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:21:59.207Z",
+ "finished_at": "2026-09-19T20:21:59.226Z",
+ "queued_at": "2026-09-19T20:21:59.005Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 100,
+ "cost": 0,
+ "started_at": "2026-09-19T20:21:59.058Z",
+ "finished_at": "2026-09-19T20:23:39.053Z",
+ "queued_at": "2026-09-19T20:21:59.025Z",
+ "bytes": 6744,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-19T20:23:40.891Z"
+ },
+ {
+ "id": "7a54837c09c4",
+ "title": "Finance Competition #11 — Meme-Stock Chart With a Personality",
+ "prompt": "Build this as one complete, self-contained HTML file: An INVENTED meme-stock candlestick chart that reacts to a 'hype' slider (rockets, panic), draws a moving average, shows a crosshair readout and volume, and has a fake pundit narrate the big moves. 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": "2026-09-19T20:28:43.145Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "error",
+ "error": "timeout after 600s",
+ "seconds": 600,
+ "cost": null,
+ "started_at": "2026-09-19T20:28:43.400Z",
+ "finished_at": "2026-09-19T20:38:43.561Z",
+ "queued_at": "2026-09-19T20:28:43.185Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:28:43.437Z",
+ "finished_at": "2026-09-19T20:28:43.519Z",
+ "queued_at": "2026-09-19T20:28:43.220Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:28:43.476Z",
+ "finished_at": "2026-09-19T20:28:43.680Z",
+ "queued_at": "2026-09-19T20:28:43.259Z"
+ },
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-19T20:38:43.598Z",
+ "finished_at": null,
+ "queued_at": "2026-09-19T20:28:43.296Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:28:43.557Z",
+ "finished_at": "2026-09-19T20:28:43.627Z",
+ "queued_at": "2026-09-19T20:28:43.336Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T20:28:43.716Z",
+ "finished_at": "2026-09-19T20:28:43.880Z",
+ "queued_at": "2026-09-19T20:28:43.365Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index 4019be7..4752964 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":112,"compCount":{"Art":22,"Culture":21,"News":21,"Political Cartoon":21,"Games":11,"Real Work":11,"Data Viz":11,"Science":11,"Music":11,"Sports":11,"History":11,"Nature & Space":10,"Comedy":10,"Demoscene":10,"Finance":10,"Design":10},"catIdx":11,"perCat":{"Art":1,"Culture":1,"News":1,"Political Cartoon":1,"Games":2,"Real Work":2,"Data Viz":2,"Science":2,"Music":2,"Sports":2,"History":2,"Nature & Space":1,"Comedy":1,"Demoscene":1,"Finance":1,"Design":1}}
\ No newline at end of file
+{"promptIdx":3,"modelOffset":7,"compCount":{"Art":22,"Culture":21,"News":21,"Political Cartoon":21,"Games":11,"Real Work":11,"Data Viz":11,"Science":11,"Music":11,"Sports":11,"History":11,"Nature & Space":11,"Comedy":11,"Demoscene":11,"Finance":11,"Design":10},"catIdx":15,"perCat":{"Art":1,"Culture":1,"News":1,"Political Cartoon":1,"Games":2,"Real Work":2,"Data Viz":2,"Science":2,"Music":2,"Sports":2,"History":2,"Nature & Space":2,"Comedy":2,"Demoscene":2,"Finance":2,"Design":1}}
\ No newline at end of file
diff --git a/yolo/daily-log.jsonl b/yolo/daily-log.jsonl
index 2682a4f..c8a347d 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -266,3 +266,6 @@
{"ts":"2026-09-19T19:56:25.830Z","id":"1fa4cdabadd7","promptId":"synth","category":"Music","title":"Music Competition #11 — The Singing Creature Synth","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-19T20:01:34.227Z","id":"e4458e583d4d","promptId":"bracket","category":"Sports","title":"Sports Competition #11 — Mascot Clash Bracket","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-19T20:06:42.508Z","id":"62f8074d9795","promptId":"empire-map","category":"History","title":"History Competition #11 — An Empire That Breathes","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-19T20:11:50.697Z","id":"7511f1a0843c","promptId":"ecosystem","category":"Nature & Space","title":"Nature & Space Competition #11 — Terrarium in a Jar","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-19T20:16:58.873Z","id":"58121b9214db","promptId":"excuse","category":"Comedy","title":"Comedy Competition #11 — The Excuse Slot Machine","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-19T20:23:43.087Z","id":"1d0dbe6f8207","promptId":"tunnel","category":"Demoscene","title":"Demoscene Competition #11 — Wormhole to Anywhere","models":6,"done":1,"errored":5,"settled":true,"aiPick":null,"loop":true}
← e9485ea auto-data-snapshot: 2026-09-19T13:09:57 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-19T14:13:49 (3 data files) — dat ac13702 →