← back to Model Arena
auto-data-snapshot: 2026-09-18T14:54:24 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
933886f862b3eade53fe935266a86c43ada72c06 · 2026-09-18 14:55:46 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit 933886f862b3eade53fe935266a86c43ada72c06
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Fri Sep 18 14:55:46 2026 -0700
auto-data-snapshot: 2026-09-18T14:54:24 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 450 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 7 +
3 files changed, 458 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index 8c6f878..c41e4ef 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -37384,5 +37384,455 @@
"judging": false,
"aiPick": null,
"judged_at": "2026-09-18T21:22:37.630Z"
+ },
+ {
+ "id": "399d4ed8ac29",
+ "title": "Arena · Art: Neoplastic Composition Studio",
+ "prompt": "Build this as one complete, self-contained HTML file: An interactive Mondrian-style neoplastic composition generator: adjustable grid, primary-color palette, line weight, randomize, and a save-as-image action. 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.",
+ "category": "Art",
+ "designTools": false,
+ "created_at": "2026-09-18T21:32:43.759Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "done",
+ "error": null,
+ "seconds": 11,
+ "cost": 0,
+ "started_at": "2026-09-18T21:32:43.804Z",
+ "finished_at": "2026-09-18T21:32:54.901Z",
+ "queued_at": "2026-09-18T21:32:43.767Z",
+ "bytes": 1016,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:32:43.814Z",
+ "finished_at": "2026-09-18T21:32:43.845Z",
+ "queued_at": "2026-09-18T21:32:43.775Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "done",
+ "error": null,
+ "seconds": 36,
+ "cost": 0,
+ "started_at": "2026-09-18T21:32:54.916Z",
+ "finished_at": "2026-09-18T21:33:31.258Z",
+ "queued_at": "2026-09-18T21:32:43.782Z",
+ "bytes": 4374,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "exo__gemma-4-26b-a4b-it-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:32:43.850Z",
+ "finished_at": "2026-09-18T21:32:43.885Z",
+ "queued_at": "2026-09-18T21:32:43.787Z"
+ },
+ {
+ "model": "exo__qwen3-6-35b-a3b-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:32:43.891Z",
+ "finished_at": "2026-09-18T21:32:43.921Z",
+ "queued_at": "2026-09-18T21:32:43.792Z"
+ },
+ {
+ "model": "exo__qwen3-vl-4b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:32:43.928Z",
+ "finished_at": "2026-09-18T21:32:43.954Z",
+ "queued_at": "2026-09-18T21:32:43.798Z"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-18T21:33:33.302Z"
+ },
+ {
+ "id": "07baf66b1e8e",
+ "title": "Arena · Art: Kinetic Typography Poster",
+ "prompt": "Build this as one complete, self-contained HTML file: An animated typographic art poster on a strong editorial grid that reacts to the cursor with restrained, tasteful motion and respects prefers-reduced-motion. 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.",
+ "category": "Art",
+ "designTools": false,
+ "created_at": "2026-09-18T21:38:32.012Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-7",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 1,
+ "cost": null,
+ "started_at": "2026-09-18T21:38:32.070Z",
+ "finished_at": "2026-09-18T21:38:32.714Z",
+ "queued_at": "2026-09-18T21:38:32.030Z"
+ },
+ {
+ "model": "exo__llama-3-3-70b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:38:32.733Z",
+ "finished_at": "2026-09-18T21:38:33.058Z",
+ "queued_at": "2026-09-18T21:38:32.043Z"
+ },
+ {
+ "model": "exo__gemma-4-e2b-it-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:38:33.071Z",
+ "finished_at": "2026-09-18T21:38:33.296Z",
+ "queued_at": "2026-09-18T21:38:32.050Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:38:33.311Z",
+ "finished_at": "2026-09-18T21:38:33.564Z",
+ "queued_at": "2026-09-18T21:38:32.055Z"
+ },
+ {
+ "model": "exo__qwen3-6-27b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:38:33.584Z",
+ "finished_at": "2026-09-18T21:38:33.794Z",
+ "queued_at": "2026-09-18T21:38:32.060Z"
+ },
+ {
+ "model": "exo__glm-4-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:38:33.817Z",
+ "finished_at": "2026-09-18T21:38:33.997Z",
+ "queued_at": "2026-09-18T21:38:32.065Z"
+ }
+ ]
+ },
+ {
+ "id": "6a6acec61489",
+ "title": "Art Competition #1 — 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": "2026-09-18T21:42:31.875Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen25-7b",
+ "status": "done",
+ "error": null,
+ "seconds": 26,
+ "cost": 0,
+ "started_at": "2026-09-18T21:42:31.892Z",
+ "finished_at": "2026-09-18T21:42:57.611Z",
+ "queued_at": "2026-09-18T21:42:31.881Z",
+ "bytes": 3672,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 142,
+ "cost": 0,
+ "started_at": "2026-09-18T21:42:57.620Z",
+ "finished_at": "2026-09-18T21:45:19.616Z",
+ "queued_at": "2026-09-18T21:42:31.886Z",
+ "bytes": 8461,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-18T21:45:21.323Z"
+ },
+ {
+ "id": "da248405ff0b",
+ "title": "Art Competition #2 — 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": "2026-09-18T21:44:02.621Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__deepseek-v3-1-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:44:02.661Z",
+ "finished_at": "2026-09-18T21:44:02.696Z",
+ "queued_at": "2026-09-18T21:44:02.628Z"
+ },
+ {
+ "model": "exo__llama-3-1-nemotron-70b-instruct-hf-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:44:02.701Z",
+ "finished_at": "2026-09-18T21:44:02.741Z",
+ "queued_at": "2026-09-18T21:44:02.634Z"
+ },
+ {
+ "model": "exo__gemma-4-e4b-it-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:44:02.746Z",
+ "finished_at": "2026-09-18T21:44:02.776Z",
+ "queued_at": "2026-09-18T21:44:02.639Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-5bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:44:02.781Z",
+ "finished_at": "2026-09-18T21:44:02.807Z",
+ "queued_at": "2026-09-18T21:44:02.645Z"
+ },
+ {
+ "model": "exo__glm-4-7-flash-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:44:02.812Z",
+ "finished_at": "2026-09-18T21:44:02.833Z",
+ "queued_at": "2026-09-18T21:44:02.650Z"
+ },
+ {
+ "model": "exo__qwen3-5-35b-a3b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:44:02.839Z",
+ "finished_at": "2026-09-18T21:44:02.856Z",
+ "queued_at": "2026-09-18T21:44:02.656Z"
+ }
+ ]
+ },
+ {
+ "id": "4de2ecb6715d",
+ "title": "Games Competition #1 — Comet Snake in a Dying Galaxy",
+ "prompt": "Build this as one complete, self-contained HTML file: Snake, but the snake is a comet and the arena is a shrinking galaxy: eating stars grows your tail AND bends gravity, near-misses trigger slow-mo, and the high score is logged as a named 'constellation'. 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": "Games",
+ "designTools": false,
+ "created_at": "2026-09-18T21:46:43.455Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen25-7b",
+ "status": "done",
+ "error": null,
+ "seconds": 32,
+ "cost": 0,
+ "started_at": "2026-09-18T21:46:43.492Z",
+ "finished_at": "2026-09-18T21:47:15.979Z",
+ "queued_at": "2026-09-18T21:46:43.469Z",
+ "bytes": 4498,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 261,
+ "cost": 0,
+ "started_at": "2026-09-18T21:47:15.993Z",
+ "finished_at": "2026-09-18T21:51:36.766Z",
+ "queued_at": "2026-09-18T21:46:43.481Z",
+ "bytes": 14219,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-18T21:51:38.264Z"
+ },
+ {
+ "id": "6436b977da2f",
+ "title": "Culture Competition #1 — 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": "2026-09-18T21:49:10.831Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__glm-4-7-8bit-gs32",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:49:10.898Z",
+ "finished_at": "2026-09-18T21:49:10.973Z",
+ "queued_at": "2026-09-18T21:49:10.851Z"
+ },
+ {
+ "model": "exo__glm-5-1",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:49:10.980Z",
+ "finished_at": "2026-09-18T21:49:11.076Z",
+ "queued_at": "2026-09-18T21:49:10.865Z"
+ },
+ {
+ "model": "exo__kimi-k2-5",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:49:11.087Z",
+ "finished_at": "2026-09-18T21:49:11.171Z",
+ "queued_at": "2026-09-18T21:49:10.872Z"
+ },
+ {
+ "model": "exo__qwen3-5-397b-a17b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:49:11.185Z",
+ "finished_at": "2026-09-18T21:49:11.285Z",
+ "queued_at": "2026-09-18T21:49:10.879Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-5bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:49:11.305Z",
+ "finished_at": "2026-09-18T21:49:11.360Z",
+ "queued_at": "2026-09-18T21:49:10.886Z"
+ },
+ {
+ "model": "exo__qwen3-5-2b-mlx-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:49:11.374Z",
+ "finished_at": "2026-09-18T21:49:11.426Z",
+ "queued_at": "2026-09-18T21:49:10.893Z"
+ }
+ ]
+ },
+ {
+ "id": "5152f259f6ab",
+ "title": "News Competition #1 — 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": "2026-09-18T21:54:18.996Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__deepseek-v3-2-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:54:19.040Z",
+ "finished_at": "2026-09-18T21:54:19.140Z",
+ "queued_at": "2026-09-18T21:54:19.004Z"
+ },
+ {
+ "model": "exo__qwen3-5-27b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:54:19.147Z",
+ "finished_at": "2026-09-18T21:54:19.203Z",
+ "queued_at": "2026-09-18T21:54:19.010Z"
+ },
+ {
+ "model": "exo__gemma-4-e2b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:54:19.210Z",
+ "finished_at": "2026-09-18T21:54:19.272Z",
+ "queued_at": "2026-09-18T21:54:19.017Z"
+ },
+ {
+ "model": "exo__qwen3-coder-next-5bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:54:19.279Z",
+ "finished_at": "2026-09-18T21:54:19.490Z",
+ "queued_at": "2026-09-18T21:54:19.022Z"
+ },
+ {
+ "model": "exo__glm-4-7-flash-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:54:19.502Z",
+ "finished_at": "2026-09-18T21:54:19.718Z",
+ "queued_at": "2026-09-18T21:54:19.028Z"
+ },
+ {
+ "model": "exo__qwen3-6-35b-a3b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-18T21:54:19.730Z",
+ "finished_at": "2026-09-18T21:54:19.858Z",
+ "queued_at": "2026-09-18T21:54:19.034Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index 301a5df..8aff66d 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":1,"modelOffset":6}
\ No newline at end of file
+{"promptIdx":3,"modelOffset":42,"compCount":{"Art":2,"Culture":1,"News":1},"catIdx":3,"perCat":{"Art":1,"Culture":1,"News":1}}
\ No newline at end of file
diff --git a/yolo/daily-log.jsonl b/yolo/daily-log.jsonl
index 5123089..60858e7 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -50,3 +50,10 @@
{"ts":"2026-09-17T14:18:09.869Z","id":"64a389aa718d","promptId":"dashboard","title":"Daily · Real Work: Operations Dashboard","done":2,"aiPick":"hermes3-8b","partial":false}
{"ts":"2026-09-18T14:16:51.457Z","id":"48ce08a3d72f","promptId":"quiz","title":"Daily · Education: Adaptive Design Quiz","done":3,"aiPick":"qwen25-7b","partial":false}
{"ts":"2026-09-18T21:22:26.598Z","id":"0f99d5cfb47a","promptId":"gen-art","category":"Art","title":"Arena · Art: Generative Art Canvas","models":2,"done":1,"errored":1,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-18T21:33:31.827Z","id":"399d4ed8ac29","promptId":"mondrian","category":"Art","title":"Arena · Art: Neoplastic Composition Studio","models":6,"done":2,"errored":4,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-18T21:38:40.073Z","id":"07baf66b1e8e","promptId":"kinetic-type","category":"Art","title":"Arena · Art: Kinetic Typography Poster","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-18T21:44:10.665Z","id":"da248405ff0b","promptId":"gen-art","category":"Art","title":"Art Competition #2 — Living Ink Canvas","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-18T21:45:19.952Z","id":"6a6acec61489","promptId":"gen-art","category":"Art","title":"Art Competition #1 — Living Ink Canvas","models":2,"done":2,"errored":0,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-18T21:49:18.902Z","id":"6436b977da2f","promptId":"museum","category":"Culture","title":"Culture Competition #1 — Museum of a Lost Civilization","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-18T21:51:39.629Z","id":"4de2ecb6715d","promptId":"snake","category":"Games","title":"Games Competition #1 — Comet Snake in a Dying Galaxy","models":2,"done":2,"errored":0,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-18T21:54:27.046Z","id":"5152f259f6ab","promptId":"election-viz","category":"News","title":"News Competition #1 — Election Night: Floating Isles","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
← 272a37d model-arena loop: 16-category creative COMPETITION bank + we
·
back to Model Arena
·
auto-data-snapshot: 2026-09-18T15:26:25 (3 data files) — dat 9524fdb →