← back to Model Arena
auto-data-snapshot: 2026-09-18T19:41:16 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
5ddedc51f3c3cc11b85648852ced71122b859d7d · 2026-09-18 19:42:17 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit 5ddedc51f3c3cc11b85648852ced71122b859d7d
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Fri Sep 18 19:42:17 2026 -0700
auto-data-snapshot: 2026-09-18T19:41:16 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 433 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 5 +
3 files changed, 439 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index 5b9baec..4953e9f 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -40330,5 +40330,438 @@
"queued_at": "2026-09-19T02:08:29.115Z"
}
]
+ },
+ {
+ "id": "a4fbc080f6b6",
+ "title": "Culture Competition #4 — 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": "2026-09-19T02:13:37.235Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__kimi-k2-thinking",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:13:37.298Z",
+ "finished_at": "2026-09-19T02:13:37.479Z",
+ "queued_at": "2026-09-19T02:13:37.248Z"
+ },
+ {
+ "model": "exo__qwen3-coder-next-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:13:37.493Z",
+ "finished_at": "2026-09-19T02:13:37.647Z",
+ "queued_at": "2026-09-19T02:13:37.260Z"
+ },
+ {
+ "model": "exo__deepseek-v3-2-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:13:37.661Z",
+ "finished_at": "2026-09-19T02:13:37.828Z",
+ "queued_at": "2026-09-19T02:13:37.272Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-nano-9b-v2-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:13:37.843Z",
+ "finished_at": "2026-09-19T02:13:37.984Z",
+ "queued_at": "2026-09-19T02:13:37.279Z"
+ },
+ {
+ "model": "exo__meta-llama-3-1-8b-instruct-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:13:38.003Z",
+ "finished_at": "2026-09-19T02:13:38.350Z",
+ "queued_at": "2026-09-19T02:13:37.285Z"
+ },
+ {
+ "model": "exo__gemma-4-e4b-it-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 1,
+ "cost": null,
+ "started_at": "2026-09-19T02:13:38.366Z",
+ "finished_at": "2026-09-19T02:13:38.908Z",
+ "queued_at": "2026-09-19T02:13:37.291Z"
+ }
+ ]
+ },
+ {
+ "id": "2cc46c42ab68",
+ "title": "News Competition #4 — 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": "2026-09-19T02:18:45.479Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__qwen3-5-397b-a17b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:18:45.565Z",
+ "finished_at": "2026-09-19T02:18:45.736Z",
+ "queued_at": "2026-09-19T02:18:45.506Z"
+ },
+ {
+ "model": "exo__glm-4-5-air-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:18:45.749Z",
+ "finished_at": "2026-09-19T02:18:45.939Z",
+ "queued_at": "2026-09-19T02:18:45.522Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:18:45.961Z",
+ "finished_at": "2026-09-19T02:18:46.315Z",
+ "queued_at": "2026-09-19T02:18:45.533Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-nvfp4",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:18:46.321Z",
+ "finished_at": "2026-09-19T02:18:46.383Z",
+ "queued_at": "2026-09-19T02:18:45.541Z"
+ },
+ {
+ "model": "exo__qwen3-6-27b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:18:46.390Z",
+ "finished_at": "2026-09-19T02:18:46.476Z",
+ "queued_at": "2026-09-19T02:18:45.550Z"
+ },
+ {
+ "model": "exo__llama-3-3-70b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:18:46.485Z",
+ "finished_at": "2026-09-19T02:18:46.829Z",
+ "queued_at": "2026-09-19T02:18:45.558Z"
+ }
+ ]
+ },
+ {
+ "id": "0fb5d35efc60",
+ "title": "Political Cartoon Competition #4 — How a Law Gets Passed (Rube Goldberg)",
+ "prompt": "Build this as one complete, self-contained HTML file: A cheerful SVG cartoon explaining how an INVENTED law gets passed as a Rube-Goldberg contraption: click to send a marble through each stage of 'the process', with captions and a gleeful finale. 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": "2026-09-19T02:23:53.842Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__glm-5-1-mxfp4-q8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:23:53.899Z",
+ "finished_at": "2026-09-19T02:23:53.934Z",
+ "queued_at": "2026-09-19T02:23:53.860Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:23:53.940Z",
+ "finished_at": "2026-09-19T02:23:54.003Z",
+ "queued_at": "2026-09-19T02:23:53.874Z"
+ },
+ {
+ "model": "exo__minimax-m2-5-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:23:54.009Z",
+ "finished_at": "2026-09-19T02:23:54.034Z",
+ "queued_at": "2026-09-19T02:23:53.879Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:23:54.040Z",
+ "finished_at": "2026-09-19T02:23:54.091Z",
+ "queued_at": "2026-09-19T02:23:53.884Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:23:54.102Z",
+ "finished_at": "2026-09-19T02:23:54.159Z",
+ "queued_at": "2026-09-19T02:23:53.889Z"
+ },
+ {
+ "model": "exo__kimi-k2-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:23:54.171Z",
+ "finished_at": "2026-09-19T02:23:54.208Z",
+ "queued_at": "2026-09-19T02:23:53.894Z"
+ }
+ ]
+ },
+ {
+ "id": "151e740be516",
+ "title": "Art Competition #6 — Living Ink Canvas",
+ "prompt": "Build this as one complete, self-contained HTML file: A generative 'living ink' canvas where cursor speed is brush pressure: fast strokes splatter, slow strokes bloom into calligraphic petals. The palette drifts with an imagined time-of-day, and one keypress freezes the piece into a framed 'gallery print' with an invented placard. 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": "2026-09-19T02:29:01.934Z",
+ "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": "2026-09-19T02:29:01.999Z",
+ "finished_at": "2026-09-19T02:29:02.013Z",
+ "queued_at": "2026-09-19T02:29:01.954Z"
+ },
+ {
+ "model": "exo__deepseek-v4-flash",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:29:02.018Z",
+ "finished_at": "2026-09-19T02:29:02.027Z",
+ "queued_at": "2026-09-19T02:29:01.965Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-thinking-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:29:02.032Z",
+ "finished_at": "2026-09-19T02:29:02.040Z",
+ "queued_at": "2026-09-19T02:29:01.972Z"
+ },
+ {
+ "model": "exo__deepseek-v4-pro",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:29:02.045Z",
+ "finished_at": "2026-09-19T02:29:02.054Z",
+ "queued_at": "2026-09-19T02:29:01.982Z"
+ },
+ {
+ "model": "exo__deepseek-v3-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:29:02.059Z",
+ "finished_at": "2026-09-19T02:29:02.069Z",
+ "queued_at": "2026-09-19T02:29:01.987Z"
+ },
+ {
+ "model": "exo__glm-5-8bit-mxfp8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:29:02.074Z",
+ "finished_at": "2026-09-19T02:29:02.085Z",
+ "queued_at": "2026-09-19T02:29:01.994Z"
+ }
+ ]
+ },
+ {
+ "id": "a339bcf33204",
+ "title": "Culture Competition #5 — Museum of a Lost Civilization",
+ "prompt": "Build this as one complete, self-contained HTML file: A pocket museum for a civilization that never existed: wander a floor-plan of rooms, click artifacts to reveal INVENTED lore, and assemble a shard-puzzle that unlocks the civilization's 'lost anthem' (Web Audio). 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": "2026-09-19T02:34:10.041Z",
+ "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": "2026-09-19T02:34:10.102Z",
+ "finished_at": "2026-09-19T02:34:10.205Z",
+ "queued_at": "2026-09-19T02:34:10.059Z"
+ },
+ {
+ "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": "2026-09-19T02:34:10.219Z",
+ "finished_at": "2026-09-19T02:34:10.256Z",
+ "queued_at": "2026-09-19T02:34:10.073Z"
+ },
+ {
+ "model": "exo__kimi-k2-6",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:34:10.266Z",
+ "finished_at": "2026-09-19T02:34:10.296Z",
+ "queued_at": "2026-09-19T02:34:10.079Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:34:10.302Z",
+ "finished_at": "2026-09-19T02:34:10.328Z",
+ "queued_at": "2026-09-19T02:34:10.086Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:34:10.333Z",
+ "finished_at": "2026-09-19T02:34:10.364Z",
+ "queued_at": "2026-09-19T02:34:10.091Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 156,
+ "cost": 0,
+ "started_at": "2026-09-19T02:34:10.108Z",
+ "finished_at": "2026-09-19T02:36:45.903Z",
+ "queued_at": "2026-09-19T02:34:10.097Z",
+ "bytes": 10801,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-19T02:36:47.922Z"
+ },
+ {
+ "id": "42a46ff18e1b",
+ "title": "News Competition #5 — Election Night: Floating Isles",
+ "prompt": "Build this as one complete, self-contained HTML file: An election-night 'war room' for an INVENTED nation of floating islands: a live map colors in as returns land, a needle swings, invented pundits drop one-liners, and a 'what if' slider re-runs the result. 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": "2026-09-19T02:41:50.306Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-19T02:41:50.364Z",
+ "finished_at": null,
+ "queued_at": "2026-09-19T02:41:50.321Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:41:50.370Z",
+ "finished_at": "2026-09-19T02:41:50.392Z",
+ "queued_at": "2026-09-19T02:41:50.333Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-19T02:41:50.375Z",
+ "finished_at": null,
+ "queued_at": "2026-09-19T02:41:50.341Z"
+ },
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-19T02:41:50.347Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-19T02:41:50.397Z",
+ "finished_at": "2026-09-19T02:41:50.413Z",
+ "queued_at": "2026-09-19T02:41:50.353Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-19T02:41:50.358Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index 516bef6..692d1be 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":100,"compCount":{"Art":5,"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":2},"catIdx":17,"perCat":{"Art":4,"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":2}}
\ No newline at end of file
+{"promptIdx":3,"modelOffset":7,"compCount":{"Art":6,"Culture":5,"News":5,"Political Cartoon":4,"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},"catIdx":3,"perCat":{"Art":1,"Culture":1,"News":1,"Political Cartoon":4,"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 3a3a2f6..7e8c19e 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -91,3 +91,8 @@
{"ts":"2026-09-19T01:58:20.634Z","id":"af57ba933253","promptId":"candles","category":"Finance","title":"Finance Competition #2 — Meme-Stock Chart With a Personality","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-19T02:03:28.790Z","id":"1d485e9f02df","promptId":"tokens","category":"Design","title":"Design Competition #2 — Palette & Type Mad-Lab","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-19T02:08:37.127Z","id":"ad08713406be","promptId":"fractal","category":"Art","title":"Art Competition #5 — Deep-Sea Fractal Dive","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-19T02:13:45.302Z","id":"a4fbc080f6b6","promptId":"heritage","category":"Culture","title":"Culture Competition #4 — Lantern Temple by Night","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-19T02:18:53.571Z","id":"2cc46c42ab68","promptId":"climate-story","category":"News","title":"News Competition #4 — The City That Floods As You Scroll","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-19T02:24:01.902Z","id":"0fb5d35efc60","promptId":"civic-explainer","category":"Political Cartoon","title":"Political Cartoon Competition #4 — How a Law Gets Passed (Rube Goldberg)","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-19T02:29:10.004Z","id":"151e740be516","promptId":"gen-art","category":"Art","title":"Art Competition #6 — Living Ink Canvas","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-19T02:36:50.144Z","id":"a339bcf33204","promptId":"museum","category":"Culture","title":"Culture Competition #5 — Museum of a Lost Civilization","models":6,"done":1,"errored":5,"settled":true,"aiPick":null,"loop":true}
← 58a3b85 auto-data-snapshot: 2026-09-18T19:09:31 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-18T20:12:47 (3 data files) — dat a45dd23 →