← back to Model Arena
auto-data-snapshot: 2026-09-20T05:01:07 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
fd6c7c92cbe490871c0719b522348cd92481dc9f · 2026-09-20 05:02:50 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit fd6c7c92cbe490871c0719b522348cd92481dc9f
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Sun Sep 20 05:02:50 2026 -0700
auto-data-snapshot: 2026-09-20T05:01:07 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 362 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 4 +
3 files changed, 367 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index 48a507a..192cc03 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -63554,5 +63554,367 @@
"queued_at": "2026-09-20T11:24:06.802Z"
}
]
+ },
+ {
+ "id": "44e05845c9b9",
+ "title": "Culture Competition #37 — 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": "2026-09-20T11:29:14.948Z",
+ "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": "2026-09-20T11:29:15.279Z",
+ "finished_at": "2026-09-20T11:29:15.349Z",
+ "queued_at": "2026-09-20T11:29:15.009Z"
+ },
+ {
+ "model": "exo__qwen3-6-27b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:29:15.386Z",
+ "finished_at": "2026-09-20T11:29:15.412Z",
+ "queued_at": "2026-09-20T11:29:15.078Z"
+ },
+ {
+ "model": "exo__llama-3-3-70b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:29:15.440Z",
+ "finished_at": "2026-09-20T11:29:15.483Z",
+ "queued_at": "2026-09-20T11:29:15.135Z"
+ },
+ {
+ "model": "exo__glm-5-1-mxfp4-q8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:29:15.523Z",
+ "finished_at": "2026-09-20T11:29:15.571Z",
+ "queued_at": "2026-09-20T11:29:15.177Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:29:15.606Z",
+ "finished_at": "2026-09-20T11:29:15.642Z",
+ "queued_at": "2026-09-20T11:29:15.220Z"
+ },
+ {
+ "model": "exo__minimax-m2-5-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:29:15.959Z",
+ "finished_at": "2026-09-20T11:29:16.020Z",
+ "queued_at": "2026-09-20T11:29:15.252Z"
+ }
+ ]
+ },
+ {
+ "id": "457807d5a2dd",
+ "title": "News Competition #37 — 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": "2026-09-20T11:34:23.543Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-7-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:34:23.865Z",
+ "finished_at": "2026-09-20T11:34:23.901Z",
+ "queued_at": "2026-09-20T11:34:23.593Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:34:23.924Z",
+ "finished_at": "2026-09-20T11:34:23.963Z",
+ "queued_at": "2026-09-20T11:34:23.632Z"
+ },
+ {
+ "model": "exo__kimi-k2-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:34:23.986Z",
+ "finished_at": "2026-09-20T11:34:24.026Z",
+ "queued_at": "2026-09-20T11:34:23.690Z"
+ },
+ {
+ "model": "exo__llama-3-1-nemotron-70b-instruct-hf-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:34:24.053Z",
+ "finished_at": "2026-09-20T11:34:24.084Z",
+ "queued_at": "2026-09-20T11:34:23.757Z"
+ },
+ {
+ "model": "exo__deepseek-v4-flash",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:34:24.117Z",
+ "finished_at": "2026-09-20T11:34:24.156Z",
+ "queued_at": "2026-09-20T11:34:23.792Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-thinking-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:34:24.217Z",
+ "finished_at": "2026-09-20T11:34:24.279Z",
+ "queued_at": "2026-09-20T11:34:23.827Z"
+ }
+ ]
+ },
+ {
+ "id": "b37d5ef04e64",
+ "title": "Political Cartoon Competition #37 — The Seesaw Debate",
+ "prompt": "Build this as one complete, self-contained HTML file: An original SVG editorial cartoon about an INVENTED debate ('should robots get weekends?'): a seesaw of absurd figures the viewer can tip, with a caption that rewrites itself for each side. Non-partisan, invented symbols only. 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": "2026-09-20T11:39:31.936Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__deepseek-v4-pro",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:39:32.409Z",
+ "finished_at": "2026-09-20T11:39:32.500Z",
+ "queued_at": "2026-09-20T11:39:32.081Z"
+ },
+ {
+ "model": "exo__deepseek-v3-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:39:32.553Z",
+ "finished_at": "2026-09-20T11:39:32.626Z",
+ "queued_at": "2026-09-20T11:39:32.183Z"
+ },
+ {
+ "model": "exo__glm-5-8bit-mxfp8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:39:32.651Z",
+ "finished_at": "2026-09-20T11:39:32.741Z",
+ "queued_at": "2026-09-20T11:39:32.322Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-mxfp4",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:39:32.864Z",
+ "finished_at": "2026-09-20T11:39:32.929Z",
+ "queued_at": "2026-09-20T11:39:32.349Z"
+ },
+ {
+ "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": "2026-09-20T11:39:33.020Z",
+ "finished_at": "2026-09-20T11:39:33.090Z",
+ "queued_at": "2026-09-20T11:39:32.371Z"
+ },
+ {
+ "model": "exo__kimi-k2-6",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:39:33.109Z",
+ "finished_at": "2026-09-20T11:39:33.153Z",
+ "queued_at": "2026-09-20T11:39:32.391Z"
+ }
+ ]
+ },
+ {
+ "id": "08a1b73bf7ac",
+ "title": "Games Competition #19 — 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": "2026-09-20T11:44:40.537Z",
+ "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": "2026-09-20T11:44:41.127Z",
+ "finished_at": "2026-09-20T11:44:41.376Z",
+ "queued_at": "2026-09-20T11:44:40.729Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:44:41.429Z",
+ "finished_at": "2026-09-20T11:44:41.537Z",
+ "queued_at": "2026-09-20T11:44:40.838Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 213,
+ "cost": 0,
+ "started_at": "2026-09-20T11:44:41.163Z",
+ "finished_at": "2026-09-20T11:48:14.216Z",
+ "queued_at": "2026-09-20T11:44:40.892Z",
+ "bytes": 13666,
+ "thumb": true,
+ "aiScore": null,
+ "aiReason": "judge failed: all vision panelists failed"
+ },
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "error",
+ "error": "timeout after 600s",
+ "seconds": 600,
+ "cost": null,
+ "started_at": "2026-09-20T11:44:41.271Z",
+ "finished_at": "2026-09-20T11:54:41.383Z",
+ "queued_at": "2026-09-20T11:44:40.921Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:44:41.651Z",
+ "finished_at": "2026-09-20T11:44:41.844Z",
+ "queued_at": "2026-09-20T11:44:40.967Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:48:14.500Z",
+ "finished_at": "2026-09-20T11:48:14.813Z",
+ "queued_at": "2026-09-20T11:44:41.068Z"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-20T11:54:41.768Z"
+ },
+ {
+ "id": "751e060ad25a",
+ "title": "Real Work Competition #19 — Fishing-Dock CRM",
+ "prompt": "Build this as one complete, self-contained HTML file: A CRM where each lead is a fish on a line you drag between stages (Nibble -> Hooked -> Landed), per-stage totals shown as a catch tally, with search, quick-add, and validation. Playful but genuinely usable. 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": "Real Work",
+ "designTools": false,
+ "created_at": "2026-09-20T11:59:41.807Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-20T11:59:42.660Z",
+ "finished_at": null,
+ "queued_at": "2026-09-20T11:59:42.262Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:59:42.698Z",
+ "finished_at": "2026-09-20T11:59:42.846Z",
+ "queued_at": "2026-09-20T11:59:42.406Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:59:42.744Z",
+ "finished_at": "2026-09-20T11:59:43.071Z",
+ "queued_at": "2026-09-20T11:59:42.479Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:59:42.884Z",
+ "finished_at": "2026-09-20T11:59:42.962Z",
+ "queued_at": "2026-09-20T11:59:42.527Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:59:43.135Z",
+ "finished_at": "2026-09-20T11:59:43.366Z",
+ "queued_at": "2026-09-20T11:59:42.587Z"
+ },
+ {
+ "model": "exo__gemma-4-26b-a4b-it-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T11:59:43.021Z",
+ "finished_at": "2026-09-20T11:59:43.184Z",
+ "queued_at": "2026-09-20T11:59:42.627Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index ebf3c77..7e0389a 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":109,"compCount":{"Art":38,"Culture":36,"News":36,"Political Cartoon":36,"Games":18,"Real Work":18,"Data Viz":18,"Science":18,"Music":18,"Sports":18,"History":18,"Nature & Space":18,"Comedy":18,"Demoscene":18,"Finance":18,"Design":18},"catIdx":1,"perCat":{"Art":1,"Culture":4,"News":4,"Political Cartoon":4,"Games":3,"Real Work":3,"Data Viz":3,"Science":3,"Music":3,"Sports":3,"History":3,"Nature & Space":3,"Comedy":3,"Demoscene":3,"Finance":3,"Design":3}}
\ No newline at end of file
+{"promptIdx":3,"modelOffset":10,"compCount":{"Art":38,"Culture":37,"News":37,"Political Cartoon":37,"Games":19,"Real Work":19,"Data Viz":18,"Science":18,"Music":18,"Sports":18,"History":18,"Nature & Space":18,"Comedy":18,"Demoscene":18,"Finance":18,"Design":18},"catIdx":6,"perCat":{"Art":1,"Culture":1,"News":1,"Political Cartoon":1,"Games":1,"Real Work":1,"Data Viz":3,"Science":3,"Music":3,"Sports":3,"History":3,"Nature & Space":3,"Comedy":3,"Demoscene":3,"Finance":3,"Design":3}}
\ No newline at end of file
diff --git a/yolo/daily-log.jsonl b/yolo/daily-log.jsonl
index 61c2f19..8c71aef 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -416,3 +416,7 @@
{"ts":"2026-09-20T11:13:56.981Z","id":"c56b9bfeca30","promptId":"climate-story","category":"News","title":"News Competition #36 — The City That Floods As You Scroll","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-20T11:19:06.564Z","id":"48eff111b37c","promptId":"civic-explainer","category":"Political Cartoon","title":"Political Cartoon Competition #36 — How a Law Gets Passed (Rube Goldberg)","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-20T11:24:14.868Z","id":"a303c47677a3","promptId":"gen-art","category":"Art","title":"Art Competition #38 — Living Ink Canvas","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T11:29:23.285Z","id":"44e05845c9b9","promptId":"museum","category":"Culture","title":"Culture Competition #37 — Museum of a Lost Civilization","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T11:34:31.871Z","id":"457807d5a2dd","promptId":"election-viz","category":"News","title":"News Competition #37 — Election Night: Floating Isles","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T11:39:40.415Z","id":"b37d5ef04e64","promptId":"editorial-cartoon","category":"Political Cartoon","title":"Political Cartoon Competition #37 — The Seesaw Debate","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T11:54:41.698Z","id":"08a1b73bf7ac","promptId":"snake","category":"Games","title":"Games Competition #19 — Comet Snake in a Dying Galaxy","models":6,"done":1,"errored":5,"settled":true,"aiPick":null,"loop":true}
← 532833c auto-data-snapshot: 2026-09-20T04:26:57 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-20T05:34:32 (3 data files) — dat 9c870d1 →