← back to Model Arena
auto-data-snapshot: 2026-09-21T02:49:08 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
6a949b49e87504ed1c0d9b3140e954ed54d541b8 · 2026-09-21 02:50:56 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit 6a949b49e87504ed1c0d9b3140e954ed54d541b8
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Mon Sep 21 02:50:56 2026 -0700
auto-data-snapshot: 2026-09-21T02:49:08 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 357 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 4 +
3 files changed, 362 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index b89a510..1e5a76d 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -78856,5 +78856,362 @@
"queued_at": "2026-09-21T09:17:36.641Z"
}
]
+ },
+ {
+ "id": "9b63864b1fce",
+ "title": "Design Competition #29 — Palette & Type Mad-Lab",
+ "prompt": "Build this as one complete, self-contained HTML file: A palette + type 'mad scientist lab' where you mix two seed colors and a mood to brew a full token set, preview it live on a sample product card, copy the CSS variables, and hit 'randomize a happy accident'. 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": "2026-09-21T09:22:44.697Z",
+ "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": "2026-09-21T09:22:44.930Z",
+ "finished_at": "2026-09-21T09:22:44.961Z",
+ "queued_at": "2026-09-21T09:22:44.741Z"
+ },
+ {
+ "model": "exo__meta-llama-3-1-8b-instruct-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:22:44.988Z",
+ "finished_at": "2026-09-21T09:22:45.018Z",
+ "queued_at": "2026-09-21T09:22:44.790Z"
+ },
+ {
+ "model": "exo__gemma-4-e4b-it-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:22:45.045Z",
+ "finished_at": "2026-09-21T09:22:45.075Z",
+ "queued_at": "2026-09-21T09:22:44.824Z"
+ },
+ {
+ "model": "exo__qwen3-5-397b-a17b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:22:45.115Z",
+ "finished_at": "2026-09-21T09:22:45.153Z",
+ "queued_at": "2026-09-21T09:22:44.851Z"
+ },
+ {
+ "model": "exo__glm-4-5-air-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:22:45.176Z",
+ "finished_at": "2026-09-21T09:22:45.208Z",
+ "queued_at": "2026-09-21T09:22:44.876Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:22:45.234Z",
+ "finished_at": "2026-09-21T09:22:45.265Z",
+ "queued_at": "2026-09-21T09:22:44.902Z"
+ }
+ ]
+ },
+ {
+ "id": "01d229eecf9e",
+ "title": "Art Competition #59 — 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": "2026-09-21T09:27:52.956Z",
+ "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": "2026-09-21T09:27:53.153Z",
+ "finished_at": "2026-09-21T09:27:53.205Z",
+ "queued_at": "2026-09-21T09:27:52.984Z"
+ },
+ {
+ "model": "exo__qwen3-6-27b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:27:53.256Z",
+ "finished_at": "2026-09-21T09:27:53.301Z",
+ "queued_at": "2026-09-21T09:27:53.010Z"
+ },
+ {
+ "model": "exo__llama-3-3-70b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:27:53.326Z",
+ "finished_at": "2026-09-21T09:27:53.357Z",
+ "queued_at": "2026-09-21T09:27:53.035Z"
+ },
+ {
+ "model": "exo__glm-5-1-mxfp4-q8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:27:53.381Z",
+ "finished_at": "2026-09-21T09:27:53.409Z",
+ "queued_at": "2026-09-21T09:27:53.059Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:27:53.433Z",
+ "finished_at": "2026-09-21T09:27:53.463Z",
+ "queued_at": "2026-09-21T09:27:53.082Z"
+ },
+ {
+ "model": "exo__minimax-m2-5-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:27:53.488Z",
+ "finished_at": "2026-09-21T09:27:53.526Z",
+ "queued_at": "2026-09-21T09:27:53.117Z"
+ }
+ ]
+ },
+ {
+ "id": "cb2896bae75d",
+ "title": "Culture Competition #58 — 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": "2026-09-21T09:33:01.189Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-7-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:33:01.429Z",
+ "finished_at": "2026-09-21T09:33:01.459Z",
+ "queued_at": "2026-09-21T09:33:01.240Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:33:01.485Z",
+ "finished_at": "2026-09-21T09:33:01.520Z",
+ "queued_at": "2026-09-21T09:33:01.281Z"
+ },
+ {
+ "model": "exo__kimi-k2-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:33:01.545Z",
+ "finished_at": "2026-09-21T09:33:01.574Z",
+ "queued_at": "2026-09-21T09:33:01.320Z"
+ },
+ {
+ "model": "exo__llama-3-1-nemotron-70b-instruct-hf-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:33:01.597Z",
+ "finished_at": "2026-09-21T09:33:01.624Z",
+ "queued_at": "2026-09-21T09:33:01.347Z"
+ },
+ {
+ "model": "exo__deepseek-v4-flash",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:33:01.649Z",
+ "finished_at": "2026-09-21T09:33:01.677Z",
+ "queued_at": "2026-09-21T09:33:01.381Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-thinking-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:33:01.702Z",
+ "finished_at": "2026-09-21T09:33:01.738Z",
+ "queued_at": "2026-09-21T09:33:01.406Z"
+ }
+ ]
+ },
+ {
+ "id": "be3ec3fcc61b",
+ "title": "News Competition #58 — 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": "2026-09-21T09:38:09.466Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__deepseek-v4-pro",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:38:09.859Z",
+ "finished_at": "2026-09-21T09:38:09.910Z",
+ "queued_at": "2026-09-21T09:38:09.525Z"
+ },
+ {
+ "model": "exo__deepseek-v3-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:38:09.948Z",
+ "finished_at": "2026-09-21T09:38:09.981Z",
+ "queued_at": "2026-09-21T09:38:09.592Z"
+ },
+ {
+ "model": "exo__glm-5-8bit-mxfp8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:38:10.019Z",
+ "finished_at": "2026-09-21T09:38:10.055Z",
+ "queued_at": "2026-09-21T09:38:09.659Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-mxfp4",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:38:10.086Z",
+ "finished_at": "2026-09-21T09:38:10.144Z",
+ "queued_at": "2026-09-21T09:38:09.708Z"
+ },
+ {
+ "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": "2026-09-21T09:38:10.193Z",
+ "finished_at": "2026-09-21T09:38:10.239Z",
+ "queued_at": "2026-09-21T09:38:09.767Z"
+ },
+ {
+ "model": "exo__kimi-k2-6",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:38:10.282Z",
+ "finished_at": "2026-09-21T09:38:10.338Z",
+ "queued_at": "2026-09-21T09:38:09.805Z"
+ }
+ ]
+ },
+ {
+ "id": "e5f0df50ebb9",
+ "title": "Political Cartoon Competition #58 — 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": "2026-09-21T09:43:17.892Z",
+ "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": "2026-09-21T09:43:18.098Z",
+ "finished_at": "2026-09-21T09:43:18.183Z",
+ "queued_at": "2026-09-21T09:43:17.920Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:43:18.208Z",
+ "finished_at": "2026-09-21T09:43:18.236Z",
+ "queued_at": "2026-09-21T09:43:17.947Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 123,
+ "cost": 0,
+ "started_at": "2026-09-21T09:43:18.125Z",
+ "finished_at": "2026-09-21T09:45:20.969Z",
+ "queued_at": "2026-09-21T09:43:17.972Z",
+ "bytes": 7463,
+ "thumb": true
+ },
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-21T09:43:18.153Z",
+ "finished_at": null,
+ "queued_at": "2026-09-21T09:43:18.005Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:43:18.260Z",
+ "finished_at": "2026-09-21T09:43:18.287Z",
+ "queued_at": "2026-09-21T09:43:18.032Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T09:45:21.003Z",
+ "finished_at": "2026-09-21T09:45:21.198Z",
+ "queued_at": "2026-09-21T09:43:18.063Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index 2d5917e..dd680f1 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":103,"compCount":{"Art":58,"Culture":57,"News":57,"Political Cartoon":57,"Games":29,"Real Work":29,"Data Viz":29,"Science":29,"Music":29,"Sports":29,"History":29,"Nature & Space":29,"Comedy":29,"Demoscene":29,"Finance":29,"Design":28},"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
+{"promptIdx":3,"modelOffset":4,"compCount":{"Art":59,"Culture":58,"News":58,"Political Cartoon":58,"Games":29,"Real Work":29,"Data Viz":29,"Science":29,"Music":29,"Sports":29,"History":29,"Nature & Space":29,"Comedy":29,"Demoscene":29,"Finance":29,"Design":29},"catIdx":0,"perCat":{"Art":2,"Culture":2,"News":2,"Political Cartoon":2,"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":2}}
\ No newline at end of file
diff --git a/yolo/daily-log.jsonl b/yolo/daily-log.jsonl
index 4d8a0d4..d48efb9 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -631,3 +631,7 @@
{"ts":"2026-09-21T09:07:27.882Z","id":"d4a6c555305e","promptId":"excuse","category":"Comedy","title":"Comedy Competition #29 — The Excuse Slot Machine","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-21T09:12:36.432Z","id":"bb0b8e329d81","promptId":"tunnel","category":"Demoscene","title":"Demoscene Competition #29 — Wormhole to Anywhere","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-21T09:17:44.670Z","id":"e38ac8107886","promptId":"candles","category":"Finance","title":"Finance Competition #29 — Meme-Stock Chart With a Personality","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T09:22:52.937Z","id":"9b63864b1fce","promptId":"tokens","category":"Design","title":"Design Competition #29 — Palette & Type Mad-Lab","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T09:28:01.160Z","id":"01d229eecf9e","promptId":"mondrian","category":"Art","title":"Art Competition #59 — Playable Mondrian","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T09:33:09.437Z","id":"cb2896bae75d","promptId":"culture-map","category":"Culture","title":"Culture Competition #58 — Festivals of Nowhere","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T09:38:17.864Z","id":"be3ec3fcc61b","promptId":"econ-brief","category":"News","title":"News Competition #58 — Teletext Econ Briefing","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
← 2d10300 auto-data-snapshot: 2026-09-21T02:15:57 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-21T03:22:12 (3 data files) — dat b71d3e2 →