← back to Model Arena
auto-data-snapshot: 2026-09-26T08:50:16 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
b0b30add16a63e0c106aae999512dbdde99e295e · 2026-09-26 08:51:15 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit b0b30add16a63e0c106aae999512dbdde99e295e
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Sat Sep 26 08:51:15 2026 -0700
auto-data-snapshot: 2026-09-26T08:50:16 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 251 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 2 +
3 files changed, 254 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index f64b8de..6cc3119 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -94057,5 +94057,256 @@
}
],
"aiPick": null
+ },
+ {
+ "id": "53923fadf112",
+ "title": "History Competition #38 — An Empire That Breathes",
+ "prompt": "Build this as one complete, self-contained HTML file: An INVENTED empire that breathes across a map as you drag a century slider — borders bloom and recede, cities rise as glowing dots, and event notes fade in at the turning points, with a legend. 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": "History",
+ "designTools": false,
+ "created_at": "2026-09-26T15:23:15.061Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "timeout after 300s",
+ "seconds": 300,
+ "cost": null,
+ "started_at": "2026-09-26T15:23:15.333Z",
+ "finished_at": "2026-09-26T15:28:15.668Z",
+ "queued_at": "2026-09-26T15:23:15.099Z",
+ "fellBack": "http://127.0.0.1:8000 (mlx-community/Qwen3-30B-A3B-Instruct-2507-4bit)"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "done",
+ "error": null,
+ "seconds": 33,
+ "cost": 0,
+ "started_at": "2026-09-26T15:23:15.363Z",
+ "finished_at": "2026-09-26T15:23:48.499Z",
+ "queued_at": "2026-09-26T15:23:15.149Z",
+ "bytes": 4812
+ },
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 300,
+ "cost": null,
+ "started_at": "2026-09-26T15:23:15.389Z",
+ "finished_at": "2026-09-26T15:28:15.525Z",
+ "queued_at": "2026-09-26T15:23:15.194Z",
+ "fellBack": "http://127.0.0.1:52415 (mlx-community/Qwen3-Next-80B-A3B-Instruct-4bit)"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-4bit",
+ "status": "error",
+ "error": "timeout after 300s",
+ "seconds": 300,
+ "cost": null,
+ "started_at": "2026-09-26T15:28:15.718Z",
+ "finished_at": "2026-09-26T15:33:15.998Z",
+ "queued_at": "2026-09-26T15:23:15.238Z",
+ "fellBack": "http://127.0.0.1:8000 (mlx-community/Qwen3-30B-A3B-Instruct-2507-4bit)"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "done",
+ "error": null,
+ "seconds": 10,
+ "cost": 0,
+ "started_at": "2026-09-26T15:23:48.598Z",
+ "finished_at": "2026-09-26T15:23:58.747Z",
+ "queued_at": "2026-09-26T15:23:15.265Z",
+ "bytes": 697
+ },
+ {
+ "model": "mlx__qwen3-8-27b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 301,
+ "cost": null,
+ "started_at": "2026-09-26T15:28:15.565Z",
+ "finished_at": "2026-09-26T15:33:16.136Z",
+ "queued_at": "2026-09-26T15:23:15.292Z",
+ "fellBack": "http://127.0.0.1:52415 (mlx-community/MiniMax-M2.7-4bit)"
+ }
+ ]
+ },
+ {
+ "id": "58f3c99d2354",
+ "title": "Nature & Space Competition #38 — Terrarium in a Jar",
+ "prompt": "Build this as one complete, self-contained HTML file: A jar-terrarium ecosystem of INVENTED critters with live population graphs — feed them, introduce a predator, or trigger a drought and watch the food web wobble toward chaos or balance, with a reset. 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": "Nature & Space",
+ "designTools": false,
+ "created_at": "2026-09-26T15:38:20.087Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__qwen3-5-122b-a10b-4bit",
+ "status": "error",
+ "error": "server restarted mid-run",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-26T15:38:20.872Z",
+ "finished_at": null,
+ "queued_at": "2026-09-26T15:38:20.429Z",
+ "fellBack": "http://127.0.0.1:8000 (mlx-community/Qwen3-30B-A3B-Instruct-2507-4bit)",
+ "resumes": 3
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "done",
+ "error": null,
+ "seconds": 49,
+ "cost": 0,
+ "started_at": "2026-09-26T15:38:21.010Z",
+ "finished_at": "2026-09-26T15:39:09.829Z",
+ "queued_at": "2026-09-26T15:38:20.504Z",
+ "bytes": 6310
+ },
+ {
+ "model": "mlx__mlx-community-qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "error",
+ "error": "server restarted mid-run",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-26T15:38:21.127Z",
+ "finished_at": null,
+ "queued_at": "2026-09-26T15:38:20.638Z",
+ "resumes": 3
+ },
+ {
+ "model": "exo__gemma-4-26b-a4b-it-6bit",
+ "status": "error",
+ "error": "server restarted mid-run",
+ "seconds": 0,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-26T15:38:20.712Z",
+ "resumes": 3
+ },
+ {
+ "model": "exo__qwen3-6-35b-a3b-bf16",
+ "status": "error",
+ "error": "server restarted mid-run",
+ "seconds": 0,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-26T15:38:20.745Z",
+ "resumes": 3
+ },
+ {
+ "model": "exo__qwen3-vl-4b-instruct-4bit",
+ "status": "error",
+ "error": "server restarted mid-run",
+ "seconds": 0,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-26T15:38:20.825Z",
+ "resumes": 3
+ }
+ ],
+ "aiPick": null
+ },
+ {
+ "id": "7e115beb805b",
+ "title": "TK-12121 verify · thumb+judge path",
+ "prompt": "Build one self-contained HTML file: a centered card that says Hello with a button that toggles a color.",
+ "category": "Verification",
+ "designTools": false,
+ "created_at": "2026-09-26T15:43:40.906Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "qwen25-7b",
+ "status": "done",
+ "error": null,
+ "seconds": 14,
+ "cost": 0,
+ "started_at": "2026-09-26T15:43:41.135Z",
+ "finished_at": "2026-09-26T15:43:55.467Z",
+ "queued_at": "2026-09-26T15:43:41.051Z",
+ "bytes": 1112
+ }
+ ]
+ },
+ {
+ "id": "f630bc9a0deb",
+ "title": "Comedy Competition #38 — The Excuse Slot Machine",
+ "prompt": "Build this as one complete, self-contained HTML file: An 'Excuse Engine' slot machine that spins three reels (who / what / cosmic reason) into gloriously implausible excuses, with a believability meter, a 'too far' lever, favorites, and copy-to-clipboard. 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": "Comedy",
+ "designTools": false,
+ "created_at": "2026-09-26T15:45:48.937Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-7",
+ "status": "error",
+ "error": "timeout after 300s",
+ "seconds": 301,
+ "cost": null,
+ "started_at": "2026-09-26T15:45:50.918Z",
+ "finished_at": "2026-09-26T15:50:52.109Z",
+ "queued_at": "2026-09-26T15:45:49.251Z",
+ "fellBack": "http://127.0.0.1:8000 (mlx-community/Qwen3-30B-A3B-Instruct-2507-4bit)"
+ },
+ {
+ "model": "exo__llama-3-3-70b-instruct-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-26T15:50:52.257Z",
+ "finished_at": null,
+ "queued_at": "2026-09-26T15:45:49.869Z",
+ "fellBack": "http://127.0.0.1:8000 (mlx-community/Qwen3-30B-A3B-Instruct-2507-4bit)"
+ },
+ {
+ "model": "exo__gemma-4-e2b-it-bf16",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-26T15:45:49.924Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-instruct-8bit",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-26T15:45:50.215Z"
+ },
+ {
+ "model": "exo__qwen3-6-27b-4bit",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-26T15:45:50.352Z"
+ },
+ {
+ "model": "exo__glm-4-7-4bit",
+ "status": "queued",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": null,
+ "finished_at": null,
+ "queued_at": "2026-09-26T15:45:50.626Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index 2d89343..eedc5d6 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":2,"compCount":{"Art":76,"Culture":75,"News":75,"Political Cartoon":75,"Games":38,"Real Work":38,"Data Viz":38,"Science":38,"Music":38,"Sports":37,"History":37,"Nature & Space":37,"Comedy":37,"Demoscene":37,"Finance":37,"Design":37},"catIdx":9,"perCat":{"Art":3,"Culture":3,"News":3,"Political Cartoon":3,"Games":2,"Real Work":2,"Data Viz":2,"Science":2,"Music":2,"Sports":1,"History":1,"Nature & Space":1,"Comedy":1,"Demoscene":1,"Finance":1,"Design":1}}
\ No newline at end of file
+{"promptIdx":3,"modelOffset":20,"compCount":{"Art":76,"Culture":75,"News":75,"Political Cartoon":75,"Games":38,"Real Work":38,"Data Viz":38,"Science":38,"Music":38,"Sports":38,"History":38,"Nature & Space":38,"Comedy":38,"Demoscene":37,"Finance":37,"Design":37},"catIdx":13,"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":1,"Finance":1,"Design":1}}
\ No newline at end of file
diff --git a/yolo/daily-log.jsonl b/yolo/daily-log.jsonl
index c85028a..853b10c 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -847,3 +847,5 @@
{"ts":"2026-09-25T20:39:42.093Z","id":"833c05408dfb","promptId":"flow-infographic","category":"Data Viz","title":"Data Viz Competition #38 — Bean-to-Cup Sankey","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-25T21:16:22.944Z","id":"06da27a9a552","promptId":"pendulum","category":"Science","title":"Science Competition #38 — The Painting Pendulum","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-25T21:37:54.450Z","id":"f296e08eb20c","promptId":"synth","category":"Music","title":"Music Competition #38 — The Singing Creature Synth","models":4,"done":4,"errored":0,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-26T15:33:19.496Z","id":"53923fadf112","promptId":"empire-map","category":"History","title":"History Competition #38 — An Empire That Breathes","models":6,"done":2,"errored":4,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-26T15:40:44.983Z","id":"58f3c99d2354","promptId":"ecosystem","category":"Nature & Space","title":"Nature & Space Competition #38 — Terrarium in a Jar","models":6,"done":1,"errored":5,"settled":true,"aiPick":null,"loop":true}
← 69020db fix(TK-12121): count thumbnail by fresh PNG not Chrome exit
·
back to Model Arena
·
fix(TK-12121): exit on EADDRINUSE (single-owner guard) + thu 8d2c52a →