← back to Model Arena
auto-data-snapshot: 2026-09-20T20:39:56 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
e719e5ed4e6f3283db17fec5c834fdd7cf9909a4 · 2026-09-20 20:42:34 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit e719e5ed4e6f3283db17fec5c834fdd7cf9909a4
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Sun Sep 20 20:42:34 2026 -0700
auto-data-snapshot: 2026-09-20T20:39:56 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 213 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 3 +
3 files changed, 217 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index f337905..e4e8fa4 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -74582,5 +74582,218 @@
"judging": false,
"aiPick": null,
"judged_at": "2026-09-21T03:06:46.466Z"
+ },
+ {
+ "id": "a440a2b41a6a",
+ "title": "Design Competition #26 — 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": "2026-09-21T03:11:50.478Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "error",
+ "error": "timeout after 600s",
+ "seconds": 600,
+ "cost": null,
+ "started_at": "2026-09-21T03:11:50.737Z",
+ "finished_at": "2026-09-21T03:21:50.862Z",
+ "queued_at": "2026-09-21T03:11:50.516Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:11:50.778Z",
+ "finished_at": "2026-09-21T03:11:50.875Z",
+ "queued_at": "2026-09-21T03:11:50.552Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:11:50.818Z",
+ "finished_at": "2026-09-21T03:11:51.021Z",
+ "queued_at": "2026-09-21T03:11:50.601Z"
+ },
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "error",
+ "error": "timeout after 600s",
+ "seconds": 600,
+ "cost": null,
+ "started_at": "2026-09-21T03:21:50.898Z",
+ "finished_at": "2026-09-21T03:31:50.941Z",
+ "queued_at": "2026-09-21T03:11:50.629Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:11:50.909Z",
+ "finished_at": "2026-09-21T03:11:50.976Z",
+ "queued_at": "2026-09-21T03:11:50.659Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:11:51.048Z",
+ "finished_at": "2026-09-21T03:11:51.194Z",
+ "queued_at": "2026-09-21T03:11:50.701Z"
+ }
+ ]
+ },
+ {
+ "id": "2e4b42518ef2",
+ "title": "Art Competition #53 — 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": "2026-09-21T03:36:51.483Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__qwen3-5-122b-a10b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:36:51.836Z",
+ "finished_at": "2026-09-21T03:36:51.899Z",
+ "queued_at": "2026-09-21T03:36:51.547Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:36:51.868Z",
+ "finished_at": "2026-09-21T03:36:52.125Z",
+ "queued_at": "2026-09-21T03:36:51.610Z"
+ },
+ {
+ "model": "exo__gemma-4-26b-a4b-it-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:36:51.924Z",
+ "finished_at": "2026-09-21T03:36:51.958Z",
+ "queued_at": "2026-09-21T03:36:51.648Z"
+ },
+ {
+ "model": "exo__qwen3-6-35b-a3b-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:36:51.991Z",
+ "finished_at": "2026-09-21T03:36:52.020Z",
+ "queued_at": "2026-09-21T03:36:51.694Z"
+ },
+ {
+ "model": "exo__qwen3-vl-4b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:36:52.044Z",
+ "finished_at": "2026-09-21T03:36:52.076Z",
+ "queued_at": "2026-09-21T03:36:51.733Z"
+ },
+ {
+ "model": "exo__minimax-m2-7",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:36:52.100Z",
+ "finished_at": "2026-09-21T03:36:52.149Z",
+ "queued_at": "2026-09-21T03:36:51.786Z"
+ }
+ ]
+ },
+ {
+ "id": "682ae84662da",
+ "title": "Culture Competition #52 — 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": "2026-09-21T03:41:59.879Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__llama-3-3-70b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:42:00.202Z",
+ "finished_at": "2026-09-21T03:42:00.276Z",
+ "queued_at": "2026-09-21T03:41:59.936Z"
+ },
+ {
+ "model": "exo__gemma-4-e2b-it-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:42:00.323Z",
+ "finished_at": "2026-09-21T03:42:00.487Z",
+ "queued_at": "2026-09-21T03:41:59.980Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:42:00.518Z",
+ "finished_at": "2026-09-21T03:42:00.556Z",
+ "queued_at": "2026-09-21T03:42:00.023Z"
+ },
+ {
+ "model": "exo__qwen3-6-27b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:42:00.588Z",
+ "finished_at": "2026-09-21T03:42:00.625Z",
+ "queued_at": "2026-09-21T03:42:00.060Z"
+ },
+ {
+ "model": "exo__glm-4-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:42:00.656Z",
+ "finished_at": "2026-09-21T03:42:00.702Z",
+ "queued_at": "2026-09-21T03:42:00.092Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T03:42:00.744Z",
+ "finished_at": "2026-09-21T03:42:00.784Z",
+ "queued_at": "2026-09-21T03:42:00.136Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index 5f65ffa..eeca909 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":1,"compCount":{"Art":52,"Culture":51,"News":51,"Political Cartoon":51,"Games":26,"Real Work":26,"Data Viz":26,"Science":26,"Music":26,"Sports":26,"History":26,"Nature & Space":26,"Comedy":26,"Demoscene":26,"Finance":26,"Design":25},"catIdx":15,"perCat":{"Art":3,"Culture":3,"News":3,"Political Cartoon":3,"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":19,"compCount":{"Art":53,"Culture":52,"News":51,"Political Cartoon":51,"Games":26,"Real Work":26,"Data Viz":26,"Science":26,"Music":26,"Sports":26,"History":26,"Nature & Space":26,"Comedy":26,"Demoscene":26,"Finance":26,"Design":26},"catIdx":18,"perCat":{"Art":4,"Culture":4,"News":3,"Political Cartoon":3,"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 ae270ab..9d48bcf 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -571,3 +571,6 @@
{"ts":"2026-09-21T02:53:37.508Z","id":"fa62c2e13e2e","promptId":"excuse","category":"Comedy","title":"Comedy Competition #26 — The Excuse Slot Machine","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-21T02:58:46.005Z","id":"7f3e6aa05f45","promptId":"tunnel","category":"Demoscene","title":"Demoscene Competition #26 — Wormhole to Anywhere","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-21T03:06:50.432Z","id":"a510216ba4d8","promptId":"candles","category":"Finance","title":"Finance Competition #26 — Meme-Stock Chart With a Personality","models":6,"done":1,"errored":5,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T03:31:51.429Z","id":"a440a2b41a6a","promptId":"tokens","category":"Design","title":"Design Competition #26 — Palette & Type Mad-Lab","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T03:36:59.841Z","id":"2e4b42518ef2","promptId":"fractal","category":"Art","title":"Art Competition #53 — Deep-Sea Fractal Dive","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T03:42:08.214Z","id":"682ae84662da","promptId":"heritage","category":"Culture","title":"Culture Competition #52 — Lantern Temple by Night","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
← f792599 auto-data-snapshot: 2026-09-20T20:06:42 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-20T21:13:53 (3 data files) — dat e5babd5 →