← back to Model Arena
auto-data-snapshot: 2026-09-20T22:21:09 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
0f3735f33f1ad49d4a400984486a53ba96c6a562 · 2026-09-20 22:22:59 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit 0f3735f33f1ad49d4a400984486a53ba96c6a562
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Sun Sep 20 22:22:59 2026 -0700
auto-data-snapshot: 2026-09-20T22:21:09 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 428 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 5 +
3 files changed, 434 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index daebe5b..55dea9f 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -75718,5 +75718,433 @@
"queued_at": "2026-09-21T04:48:48.419Z"
}
]
+ },
+ {
+ "id": "72ff0bd17635",
+ "title": "Nature & Space Competition #27 — 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": "2026-09-21T04:53:56.500Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:53:56.805Z",
+ "finished_at": "2026-09-21T04:53:56.834Z",
+ "queued_at": "2026-09-21T04:53:56.557Z"
+ },
+ {
+ "model": "exo__gemma-4-e2b-it-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:53:56.861Z",
+ "finished_at": "2026-09-21T04:53:56.898Z",
+ "queued_at": "2026-09-21T04:53:56.601Z"
+ },
+ {
+ "model": "exo__qwen3-5-27b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:53:56.933Z",
+ "finished_at": "2026-09-21T04:53:57.000Z",
+ "queued_at": "2026-09-21T04:53:56.650Z"
+ },
+ {
+ "model": "exo__kimi-k2-thinking",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:53:57.030Z",
+ "finished_at": "2026-09-21T04:53:57.070Z",
+ "queued_at": "2026-09-21T04:53:56.690Z"
+ },
+ {
+ "model": "exo__qwen3-coder-next-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:53:57.113Z",
+ "finished_at": "2026-09-21T04:53:57.159Z",
+ "queued_at": "2026-09-21T04:53:56.741Z"
+ },
+ {
+ "model": "exo__deepseek-v3-2-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:53:57.201Z",
+ "finished_at": "2026-09-21T04:53:57.241Z",
+ "queued_at": "2026-09-21T04:53:56.780Z"
+ }
+ ]
+ },
+ {
+ "id": "a34eb7ab4d89",
+ "title": "Comedy Competition #27 — 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": "2026-09-21T04:59:04.860Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__nvidia-nemotron-nano-9b-v2-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:59:05.099Z",
+ "finished_at": "2026-09-21T04:59:05.153Z",
+ "queued_at": "2026-09-21T04:59:04.897Z"
+ },
+ {
+ "model": "exo__meta-llama-3-1-8b-instruct-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:59:05.200Z",
+ "finished_at": "2026-09-21T04:59:05.235Z",
+ "queued_at": "2026-09-21T04:59:04.929Z"
+ },
+ {
+ "model": "exo__gemma-4-e4b-it-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:59:05.267Z",
+ "finished_at": "2026-09-21T04:59:05.315Z",
+ "queued_at": "2026-09-21T04:59:04.964Z"
+ },
+ {
+ "model": "exo__qwen3-5-397b-a17b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:59:05.347Z",
+ "finished_at": "2026-09-21T04:59:05.390Z",
+ "queued_at": "2026-09-21T04:59:04.999Z"
+ },
+ {
+ "model": "exo__glm-4-5-air-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:59:05.432Z",
+ "finished_at": "2026-09-21T04:59:05.465Z",
+ "queued_at": "2026-09-21T04:59:05.032Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T04:59:05.503Z",
+ "finished_at": "2026-09-21T04:59:05.535Z",
+ "queued_at": "2026-09-21T04:59:05.067Z"
+ }
+ ]
+ },
+ {
+ "id": "9deca7cced0a",
+ "title": "Demoscene Competition #27 — 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": "2026-09-21T05:04:13.149Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-nvfp4",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:04:13.335Z",
+ "finished_at": "2026-09-21T05:04:13.365Z",
+ "queued_at": "2026-09-21T05:04:13.180Z"
+ },
+ {
+ "model": "exo__qwen3-6-27b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:04:13.412Z",
+ "finished_at": "2026-09-21T05:04:13.445Z",
+ "queued_at": "2026-09-21T05:04:13.203Z"
+ },
+ {
+ "model": "exo__llama-3-3-70b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:04:13.471Z",
+ "finished_at": "2026-09-21T05:04:13.501Z",
+ "queued_at": "2026-09-21T05:04:13.229Z"
+ },
+ {
+ "model": "exo__glm-5-1-mxfp4-q8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:04:13.542Z",
+ "finished_at": "2026-09-21T05:04:13.583Z",
+ "queued_at": "2026-09-21T05:04:13.264Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:04:13.606Z",
+ "finished_at": "2026-09-21T05:04:13.646Z",
+ "queued_at": "2026-09-21T05:04:13.292Z"
+ },
+ {
+ "model": "exo__minimax-m2-5-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:04:13.679Z",
+ "finished_at": "2026-09-21T05:04:13.711Z",
+ "queued_at": "2026-09-21T05:04:13.314Z"
+ }
+ ]
+ },
+ {
+ "id": "7a7abd2497f9",
+ "title": "Finance Competition #27 — 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": "2026-09-21T05:09:21.382Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-7-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:09:21.613Z",
+ "finished_at": "2026-09-21T05:09:21.645Z",
+ "queued_at": "2026-09-21T05:09:21.430Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:09:21.668Z",
+ "finished_at": "2026-09-21T05:09:21.700Z",
+ "queued_at": "2026-09-21T05:09:21.461Z"
+ },
+ {
+ "model": "exo__kimi-k2-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:09:21.733Z",
+ "finished_at": "2026-09-21T05:09:21.763Z",
+ "queued_at": "2026-09-21T05:09:21.489Z"
+ },
+ {
+ "model": "exo__llama-3-1-nemotron-70b-instruct-hf-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:09:21.789Z",
+ "finished_at": "2026-09-21T05:09:21.822Z",
+ "queued_at": "2026-09-21T05:09:21.514Z"
+ },
+ {
+ "model": "exo__deepseek-v4-flash",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:09:21.850Z",
+ "finished_at": "2026-09-21T05:09:21.877Z",
+ "queued_at": "2026-09-21T05:09:21.550Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-thinking-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:09:21.903Z",
+ "finished_at": "2026-09-21T05:09:21.932Z",
+ "queued_at": "2026-09-21T05:09:21.583Z"
+ }
+ ]
+ },
+ {
+ "id": "e8b57b4a2957",
+ "title": "Design Competition #27 — 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": "2026-09-21T05:14:29.866Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__deepseek-v4-pro",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:14:30.143Z",
+ "finished_at": "2026-09-21T05:14:30.172Z",
+ "queued_at": "2026-09-21T05:14:29.913Z"
+ },
+ {
+ "model": "exo__deepseek-v3-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:14:30.207Z",
+ "finished_at": "2026-09-21T05:14:30.248Z",
+ "queued_at": "2026-09-21T05:14:29.965Z"
+ },
+ {
+ "model": "exo__glm-5-8bit-mxfp8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:14:30.288Z",
+ "finished_at": "2026-09-21T05:14:30.324Z",
+ "queued_at": "2026-09-21T05:14:30.011Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-mxfp4",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:14:30.352Z",
+ "finished_at": "2026-09-21T05:14:30.391Z",
+ "queued_at": "2026-09-21T05:14:30.065Z"
+ },
+ {
+ "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": "2026-09-21T05:14:30.427Z",
+ "finished_at": "2026-09-21T05:14:30.480Z",
+ "queued_at": "2026-09-21T05:14:30.093Z"
+ },
+ {
+ "model": "exo__kimi-k2-6",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:14:30.507Z",
+ "finished_at": "2026-09-21T05:14:30.574Z",
+ "queued_at": "2026-09-21T05:14:30.120Z"
+ }
+ ]
+ },
+ {
+ "id": "b770c6c34b3b",
+ "title": "Art Competition #55 — 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": "2026-09-21T05:19:38.192Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:19:38.549Z",
+ "finished_at": "2026-09-21T05:19:38.663Z",
+ "queued_at": "2026-09-21T05:19:38.293Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:19:38.692Z",
+ "finished_at": "2026-09-21T05:19:38.724Z",
+ "queued_at": "2026-09-21T05:19:38.336Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 109,
+ "cost": 0,
+ "started_at": "2026-09-21T05:19:38.590Z",
+ "finished_at": "2026-09-21T05:21:28.019Z",
+ "queued_at": "2026-09-21T05:19:38.386Z",
+ "bytes": 7628,
+ "thumb": true
+ },
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-21T05:19:38.618Z",
+ "finished_at": null,
+ "queued_at": "2026-09-21T05:19:38.428Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:19:38.751Z",
+ "finished_at": "2026-09-21T05:19:38.781Z",
+ "queued_at": "2026-09-21T05:19:38.459Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T05:21:28.147Z",
+ "finished_at": "2026-09-21T05:21:28.466Z",
+ "queued_at": "2026-09-21T05:19:38.507Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index 551613e..17eb38d 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":97,"compCount":{"Art":54,"Culture":53,"News":53,"Political Cartoon":53,"Games":27,"Real Work":27,"Data Viz":27,"Science":27,"Music":27,"Sports":27,"History":27,"Nature & Space":26,"Comedy":26,"Demoscene":26,"Finance":26,"Design":26},"catIdx":11,"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":2,"Comedy":2,"Demoscene":2,"Finance":2,"Design":2}}
\ No newline at end of file
+{"promptIdx":3,"modelOffset":4,"compCount":{"Art":55,"Culture":53,"News":53,"Political Cartoon":53,"Games":27,"Real Work":27,"Data Viz":27,"Science":27,"Music":27,"Sports":27,"History":27,"Nature & Space":27,"Comedy":27,"Demoscene":27,"Finance":27,"Design":27},"catIdx":17,"perCat":{"Art":2,"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
diff --git a/yolo/daily-log.jsonl b/yolo/daily-log.jsonl
index 3be67c8..76889ed 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -587,3 +587,8 @@
{"ts":"2026-09-21T04:38:39.876Z","id":"bd7121b85384","promptId":"audioviz","category":"Music","title":"Music Competition #27 — Skyline Audio Visualizer","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-21T04:43:48.083Z","id":"0574b4d48c07","promptId":"statcard","category":"Sports","title":"Sports Competition #27 — Holographic Stat Cards","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-21T04:48:56.457Z","id":"7b4d67da0323","promptId":"artifact","category":"History","title":"History Competition #27 — Spotlit Relic Viewer","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T04:54:04.811Z","id":"72ff0bd17635","promptId":"weather","category":"Nature & Space","title":"Nature & Space Competition #27 — Shake-the-Snow-Globe Weather","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T04:59:13.102Z","id":"a34eb7ab4d89","promptId":"roast","category":"Comedy","title":"Comedy Competition #27 — Roast or Toast Campfire","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T05:04:21.340Z","id":"9deca7cced0a","promptId":"metaballs","category":"Demoscene","title":"Demoscene Competition #27 — Slime Creature Terrarium","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T05:09:29.619Z","id":"7a7abd2497f9","promptId":"budget","category":"Finance","title":"Finance Competition #27 — Budget as Jars of Coins","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T05:14:38.150Z","id":"e8b57b4a2957","promptId":"formkit","category":"Design","title":"Design Competition #27 — Pre-Flight Checklist Form Kit","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
← eeb3ebf auto-data-snapshot: 2026-09-20T21:47:38 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-20T22:54:06 (3 data files) — dat 6f9b1f5 →