← back to Model Arena
auto-data-snapshot: 2026-09-19T20:05:42 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
0976ad963a563253d2ec8a229ea91326e816d6a8 · 2026-09-19 20:07:29 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit 0976ad963a563253d2ec8a229ea91326e816d6a8
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Sat Sep 19 20:07:29 2026 -0700
auto-data-snapshot: 2026-09-19T20:05:42 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 355 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 4 +
3 files changed, 360 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index 160cccb..1e35665 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -57356,5 +57356,360 @@
"queued_at": "2026-09-20T02:33:32.986Z"
}
]
+ },
+ {
+ "id": "ed771397180b",
+ "title": "Finance Competition #14 — 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": "2026-09-20T02:38:41.231Z",
+ "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": "2026-09-20T02:38:41.378Z",
+ "finished_at": "2026-09-20T02:38:41.403Z",
+ "queued_at": "2026-09-20T02:38:41.258Z"
+ },
+ {
+ "model": "exo__meta-llama-3-1-8b-instruct-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:38:41.430Z",
+ "finished_at": "2026-09-20T02:38:41.458Z",
+ "queued_at": "2026-09-20T02:38:41.282Z"
+ },
+ {
+ "model": "exo__gemma-4-e4b-it-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:38:41.478Z",
+ "finished_at": "2026-09-20T02:38:41.500Z",
+ "queued_at": "2026-09-20T02:38:41.302Z"
+ },
+ {
+ "model": "exo__qwen3-5-397b-a17b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:38:41.522Z",
+ "finished_at": "2026-09-20T02:38:41.546Z",
+ "queued_at": "2026-09-20T02:38:41.321Z"
+ },
+ {
+ "model": "exo__glm-4-5-air-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:38:41.570Z",
+ "finished_at": "2026-09-20T02:38:41.631Z",
+ "queued_at": "2026-09-20T02:38:41.340Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:38:41.658Z",
+ "finished_at": "2026-09-20T02:38:41.689Z",
+ "queued_at": "2026-09-20T02:38:41.360Z"
+ }
+ ]
+ },
+ {
+ "id": "e08248943c0f",
+ "title": "Design Competition #14 — 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": "2026-09-20T02:43:49.401Z",
+ "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": "2026-09-20T02:43:49.558Z",
+ "finished_at": "2026-09-20T02:43:49.588Z",
+ "queued_at": "2026-09-20T02:43:49.429Z"
+ },
+ {
+ "model": "exo__qwen3-6-27b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:43:49.619Z",
+ "finished_at": "2026-09-20T02:43:49.652Z",
+ "queued_at": "2026-09-20T02:43:49.453Z"
+ },
+ {
+ "model": "exo__llama-3-3-70b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:43:49.683Z",
+ "finished_at": "2026-09-20T02:43:49.713Z",
+ "queued_at": "2026-09-20T02:43:49.474Z"
+ },
+ {
+ "model": "exo__glm-5-1-mxfp4-q8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:43:49.741Z",
+ "finished_at": "2026-09-20T02:43:49.772Z",
+ "queued_at": "2026-09-20T02:43:49.495Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:43:49.807Z",
+ "finished_at": "2026-09-20T02:43:49.850Z",
+ "queued_at": "2026-09-20T02:43:49.518Z"
+ },
+ {
+ "model": "exo__minimax-m2-5-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:43:49.888Z",
+ "finished_at": "2026-09-20T02:43:49.934Z",
+ "queued_at": "2026-09-20T02:43:49.539Z"
+ }
+ ]
+ },
+ {
+ "id": "276216e5ef95",
+ "title": "Art Competition #29 — 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": "2026-09-20T02:48:57.608Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-7-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:48:57.940Z",
+ "finished_at": "2026-09-20T02:48:57.987Z",
+ "queued_at": "2026-09-20T02:48:57.659Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:48:58.027Z",
+ "finished_at": "2026-09-20T02:48:58.067Z",
+ "queued_at": "2026-09-20T02:48:57.702Z"
+ },
+ {
+ "model": "exo__kimi-k2-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:48:58.105Z",
+ "finished_at": "2026-09-20T02:48:58.129Z",
+ "queued_at": "2026-09-20T02:48:57.751Z"
+ },
+ {
+ "model": "exo__llama-3-1-nemotron-70b-instruct-hf-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:48:58.163Z",
+ "finished_at": "2026-09-20T02:48:58.215Z",
+ "queued_at": "2026-09-20T02:48:57.802Z"
+ },
+ {
+ "model": "exo__deepseek-v4-flash",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:48:58.239Z",
+ "finished_at": "2026-09-20T02:48:58.262Z",
+ "queued_at": "2026-09-20T02:48:57.853Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-thinking-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:48:58.282Z",
+ "finished_at": "2026-09-20T02:48:58.304Z",
+ "queued_at": "2026-09-20T02:48:57.896Z"
+ }
+ ]
+ },
+ {
+ "id": "0f2faee94b69",
+ "title": "Culture Competition #28 — 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": "2026-09-20T02:54:05.970Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__deepseek-v4-pro",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:54:06.137Z",
+ "finished_at": "2026-09-20T02:54:06.166Z",
+ "queued_at": "2026-09-20T02:54:05.990Z"
+ },
+ {
+ "model": "exo__deepseek-v3-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:54:06.192Z",
+ "finished_at": "2026-09-20T02:54:06.214Z",
+ "queued_at": "2026-09-20T02:54:06.020Z"
+ },
+ {
+ "model": "exo__glm-5-8bit-mxfp8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:54:06.234Z",
+ "finished_at": "2026-09-20T02:54:06.268Z",
+ "queued_at": "2026-09-20T02:54:06.047Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-mxfp4",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:54:06.298Z",
+ "finished_at": "2026-09-20T02:54:06.331Z",
+ "queued_at": "2026-09-20T02:54:06.072Z"
+ },
+ {
+ "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": "2026-09-20T02:54:06.351Z",
+ "finished_at": "2026-09-20T02:54:06.384Z",
+ "queued_at": "2026-09-20T02:54:06.091Z"
+ },
+ {
+ "model": "exo__kimi-k2-6",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:54:06.411Z",
+ "finished_at": "2026-09-20T02:54:06.437Z",
+ "queued_at": "2026-09-20T02:54:06.115Z"
+ }
+ ]
+ },
+ {
+ "id": "5c02b79b7a7c",
+ "title": "News Competition #28 — The City That Floods As You Scroll",
+ "prompt": "Build this as one complete, self-contained HTML file: A data-journalism scrollytelling piece where an INVENTED city floods frame by frame as you drag a rising-sea slider, annotated charts surface, and a final interaction lets the reader 'plant' defenses and watch the chart bend. 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": "2026-09-20T02:59:14.177Z",
+ "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": "2026-09-20T02:59:14.355Z",
+ "finished_at": "2026-09-20T02:59:14.426Z",
+ "queued_at": "2026-09-20T02:59:14.212Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:59:14.455Z",
+ "finished_at": "2026-09-20T02:59:14.479Z",
+ "queued_at": "2026-09-20T02:59:14.242Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-20T02:59:14.373Z",
+ "finished_at": null,
+ "queued_at": "2026-09-20T02:59:14.270Z"
+ },
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-20T02:59:14.397Z",
+ "finished_at": null,
+ "queued_at": "2026-09-20T02:59:14.298Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T02:59:14.501Z",
+ "finished_at": "2026-09-20T02:59:14.535Z",
+ "queued_at": "2026-09-20T02:59:14.317Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-20T02:59:14.335Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index bd53abb..a9a4e6d 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":103,"compCount":{"Art":28,"Culture":27,"News":27,"Political Cartoon":27,"Games":14,"Real Work":14,"Data Viz":14,"Science":14,"Music":14,"Sports":14,"History":14,"Nature & Space":14,"Comedy":14,"Demoscene":14,"Finance":13,"Design":13},"catIdx":14,"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":1,"Design":1}}
\ No newline at end of file
+{"promptIdx":3,"modelOffset":4,"compCount":{"Art":29,"Culture":28,"News":28,"Political Cartoon":27,"Games":14,"Real Work":14,"Data Viz":14,"Science":14,"Music":14,"Sports":14,"History":14,"Nature & Space":14,"Comedy":14,"Demoscene":14,"Finance":14,"Design":14},"catIdx":19,"perCat":{"Art":4,"Culture":4,"News":4,"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 6b7f11b..c7153ad 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -329,3 +329,7 @@
{"ts":"2026-09-20T02:23:24.640Z","id":"d3731019c16e","promptId":"ecosystem","category":"Nature & Space","title":"Nature & Space Competition #14 — Terrarium in a Jar","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-20T02:28:32.830Z","id":"8590e61dffcf","promptId":"excuse","category":"Comedy","title":"Comedy Competition #14 — The Excuse Slot Machine","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-20T02:33:41.009Z","id":"6aafaaeb2d9b","promptId":"tunnel","category":"Demoscene","title":"Demoscene Competition #14 — Wormhole to Anywhere","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T02:38:49.383Z","id":"ed771397180b","promptId":"candles","category":"Finance","title":"Finance Competition #14 — Meme-Stock Chart With a Personality","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T02:43:57.561Z","id":"e08248943c0f","promptId":"tokens","category":"Design","title":"Design Competition #14 — Palette & Type Mad-Lab","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T02:49:05.947Z","id":"276216e5ef95","promptId":"fractal","category":"Art","title":"Art Competition #29 — Deep-Sea Fractal Dive","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T02:54:14.144Z","id":"0f2faee94b69","promptId":"heritage","category":"Culture","title":"Culture Competition #28 — Lantern Temple by Night","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
← fbd03e1 auto-data-snapshot: 2026-09-19T19:32:43 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-19T20:38:43 (3 data files) — dat 3d87f84 →