← back to Model Arena
auto-data-snapshot: 2026-09-21T11:31:47 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
75429573c816527c3a6a62476845888b0bd692f4 · 2026-09-21 11:33:33 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit 75429573c816527c3a6a62476845888b0bd692f4
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Mon Sep 21 11:33:33 2026 -0700
auto-data-snapshot: 2026-09-21T11:31:47 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 428 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 5 +
3 files changed, 434 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index cc23d82..00a346c 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -84957,5 +84957,433 @@
"queued_at": "2026-09-21T17:59:20.056Z"
}
]
+ },
+ {
+ "id": "0d720f6c2833",
+ "title": "Art Competition #68 — Swarm Typography Poster",
+ "prompt": "Build this as one complete, self-contained HTML file: A kinetic-type poster for an INVENTED avant-garde festival where the headline letters behave like a flocking swarm — scattering from the cursor and re-forming the title — with a slider that morphs the mood from 'whisper' to 'scream'. 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": "2026-09-21T18:04:28.169Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:04:28.466Z",
+ "finished_at": "2026-09-21T18:04:28.511Z",
+ "queued_at": "2026-09-21T18:04:28.217Z"
+ },
+ {
+ "model": "exo__gemma-4-e2b-it-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:04:28.550Z",
+ "finished_at": "2026-09-21T18:04:28.578Z",
+ "queued_at": "2026-09-21T18:04:28.259Z"
+ },
+ {
+ "model": "exo__qwen3-5-27b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:04:28.630Z",
+ "finished_at": "2026-09-21T18:04:28.675Z",
+ "queued_at": "2026-09-21T18:04:28.294Z"
+ },
+ {
+ "model": "exo__kimi-k2-thinking",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:04:28.720Z",
+ "finished_at": "2026-09-21T18:04:28.769Z",
+ "queued_at": "2026-09-21T18:04:28.332Z"
+ },
+ {
+ "model": "exo__qwen3-coder-next-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:04:28.817Z",
+ "finished_at": "2026-09-21T18:04:28.864Z",
+ "queued_at": "2026-09-21T18:04:28.375Z"
+ },
+ {
+ "model": "exo__deepseek-v3-2-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:04:28.913Z",
+ "finished_at": "2026-09-21T18:04:28.963Z",
+ "queued_at": "2026-09-21T18:04:28.422Z"
+ }
+ ]
+ },
+ {
+ "id": "e9e06a65bab2",
+ "title": "Culture Competition #67 — The Last Speaker",
+ "prompt": "Build this as one complete, self-contained HTML file: An interactive 'sound of a dying language' explainer for an INVENTED tongue: click words on a fading map to hear invented phonemes, watch the speaker-count clock tick down, and 'save' words into a phrasebook. 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": "2026-09-21T18:09:36.553Z",
+ "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": "2026-09-21T18:09:36.908Z",
+ "finished_at": "2026-09-21T18:09:36.971Z",
+ "queued_at": "2026-09-21T18:09:36.600Z"
+ },
+ {
+ "model": "exo__meta-llama-3-1-8b-instruct-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:09:37.022Z",
+ "finished_at": "2026-09-21T18:09:37.133Z",
+ "queued_at": "2026-09-21T18:09:36.682Z"
+ },
+ {
+ "model": "exo__gemma-4-e4b-it-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:09:37.215Z",
+ "finished_at": "2026-09-21T18:09:37.288Z",
+ "queued_at": "2026-09-21T18:09:36.722Z"
+ },
+ {
+ "model": "exo__qwen3-5-397b-a17b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:09:37.331Z",
+ "finished_at": "2026-09-21T18:09:37.405Z",
+ "queued_at": "2026-09-21T18:09:36.753Z"
+ },
+ {
+ "model": "exo__glm-4-5-air-bf16",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:09:37.446Z",
+ "finished_at": "2026-09-21T18:09:37.564Z",
+ "queued_at": "2026-09-21T18:09:36.806Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:09:37.648Z",
+ "finished_at": "2026-09-21T18:09:37.771Z",
+ "queued_at": "2026-09-21T18:09:36.846Z"
+ }
+ ]
+ },
+ {
+ "id": "d210845e524b",
+ "title": "News Competition #67 — Absurd 24-Hour News",
+ "prompt": "Build this as one complete, self-contained HTML file: A satirical INVENTED 24-hour news channel landing where three breaking stories escalate absurdly in real time via a live ticker, a 'BREAKING' klaxon toggle, and a category filter that changes the channel's whole color mood. 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": "2026-09-21T18:14:44.960Z",
+ "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": "2026-09-21T18:14:45.174Z",
+ "finished_at": "2026-09-21T18:14:45.213Z",
+ "queued_at": "2026-09-21T18:14:44.989Z"
+ },
+ {
+ "model": "exo__qwen3-6-27b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:14:45.242Z",
+ "finished_at": "2026-09-21T18:14:45.285Z",
+ "queued_at": "2026-09-21T18:14:45.018Z"
+ },
+ {
+ "model": "exo__llama-3-3-70b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:14:45.317Z",
+ "finished_at": "2026-09-21T18:14:45.352Z",
+ "queued_at": "2026-09-21T18:14:45.050Z"
+ },
+ {
+ "model": "exo__glm-5-1-mxfp4-q8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:14:45.380Z",
+ "finished_at": "2026-09-21T18:14:45.411Z",
+ "queued_at": "2026-09-21T18:14:45.078Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:14:45.442Z",
+ "finished_at": "2026-09-21T18:14:45.478Z",
+ "queued_at": "2026-09-21T18:14:45.110Z"
+ },
+ {
+ "model": "exo__minimax-m2-5-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:14:45.512Z",
+ "finished_at": "2026-09-21T18:14:45.551Z",
+ "queued_at": "2026-09-21T18:14:45.140Z"
+ }
+ ]
+ },
+ {
+ "id": "e783ea892bdb",
+ "title": "Political Cartoon Competition #67 — Ship of State by Committee",
+ "prompt": "Build this as one complete, self-contained HTML file: A single bold SVG metaphor illustration for an INVENTED political theme — 'the ship of state steered by committee' — where dragging elements changes the metaphor's meaning and the caption adapts. Light/dark toggle. 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": "2026-09-21T18:19:53.212Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-7-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:19:53.517Z",
+ "finished_at": "2026-09-21T18:19:53.574Z",
+ "queued_at": "2026-09-21T18:19:53.265Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:19:53.621Z",
+ "finished_at": "2026-09-21T18:19:53.668Z",
+ "queued_at": "2026-09-21T18:19:53.313Z"
+ },
+ {
+ "model": "exo__kimi-k2-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:19:53.714Z",
+ "finished_at": "2026-09-21T18:19:53.760Z",
+ "queued_at": "2026-09-21T18:19:53.354Z"
+ },
+ {
+ "model": "exo__llama-3-1-nemotron-70b-instruct-hf-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:19:53.811Z",
+ "finished_at": "2026-09-21T18:19:53.859Z",
+ "queued_at": "2026-09-21T18:19:53.393Z"
+ },
+ {
+ "model": "exo__deepseek-v4-flash",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:19:53.888Z",
+ "finished_at": "2026-09-21T18:19:53.916Z",
+ "queued_at": "2026-09-21T18:19:53.427Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-thinking-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:19:53.940Z",
+ "finished_at": "2026-09-21T18:19:53.969Z",
+ "queued_at": "2026-09-21T18:19:53.473Z"
+ }
+ ]
+ },
+ {
+ "id": "da2f976a90a6",
+ "title": "Games Competition #34 — 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": "2026-09-21T18:25:01.599Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__deepseek-v4-pro",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:25:01.928Z",
+ "finished_at": "2026-09-21T18:25:01.973Z",
+ "queued_at": "2026-09-21T18:25:01.647Z"
+ },
+ {
+ "model": "exo__deepseek-v3-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:25:02.001Z",
+ "finished_at": "2026-09-21T18:25:02.041Z",
+ "queued_at": "2026-09-21T18:25:01.697Z"
+ },
+ {
+ "model": "exo__glm-5-8bit-mxfp8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:25:02.067Z",
+ "finished_at": "2026-09-21T18:25:02.162Z",
+ "queued_at": "2026-09-21T18:25:01.733Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-mxfp4",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:25:02.193Z",
+ "finished_at": "2026-09-21T18:25:02.249Z",
+ "queued_at": "2026-09-21T18:25:01.777Z"
+ },
+ {
+ "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": "2026-09-21T18:25:02.279Z",
+ "finished_at": "2026-09-21T18:25:02.332Z",
+ "queued_at": "2026-09-21T18:25:01.825Z"
+ },
+ {
+ "model": "exo__kimi-k2-6",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:25:02.360Z",
+ "finished_at": "2026-09-21T18:25:02.427Z",
+ "queued_at": "2026-09-21T18:25:01.876Z"
+ }
+ ]
+ },
+ {
+ "id": "d7316b9bbdbf",
+ "title": "Real Work Competition #34 — 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": "2026-09-21T18:30:09.979Z",
+ "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": "2026-09-21T18:30:10.197Z",
+ "finished_at": "2026-09-21T18:30:10.311Z",
+ "queued_at": "2026-09-21T18:30:10.016Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:30:10.340Z",
+ "finished_at": "2026-09-21T18:30:10.371Z",
+ "queued_at": "2026-09-21T18:30:10.047Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 140,
+ "cost": 0,
+ "started_at": "2026-09-21T18:30:10.227Z",
+ "finished_at": "2026-09-21T18:32:29.820Z",
+ "queued_at": "2026-09-21T18:30:10.077Z",
+ "bytes": 9113,
+ "thumb": true
+ },
+ {
+ "model": "mlx__qwen3-30b-a3b-instruct-2507-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-21T18:30:10.255Z",
+ "finished_at": null,
+ "queued_at": "2026-09-21T18:30:10.105Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:30:10.398Z",
+ "finished_at": "2026-09-21T18:30:10.435Z",
+ "queued_at": "2026-09-21T18:30:10.132Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-21T18:32:29.902Z",
+ "finished_at": "2026-09-21T18:32:30.110Z",
+ "queued_at": "2026-09-21T18:30:10.167Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index f16dd38..3615282 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":97,"compCount":{"Art":67,"Culture":66,"News":66,"Political Cartoon":66,"Games":33,"Real Work":33,"Data Viz":33,"Science":33,"Music":33,"Sports":33,"History":33,"Nature & Space":33,"Comedy":33,"Demoscene":33,"Finance":33,"Design":33},"catIdx":0,"perCat":{"Art":2,"Culture":2,"News":2,"Political Cartoon":2,"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":4,"compCount":{"Art":68,"Culture":67,"News":67,"Political Cartoon":67,"Games":34,"Real Work":34,"Data Viz":33,"Science":33,"Music":33,"Sports":33,"History":33,"Nature & Space":33,"Comedy":33,"Demoscene":33,"Finance":33,"Design":33},"catIdx":6,"perCat":{"Art":3,"Culture":3,"News":3,"Political Cartoon":3,"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 de55348..854f232 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -717,3 +717,8 @@
{"ts":"2026-09-21T17:49:11.425Z","id":"0c4c85f8d01e","promptId":"culture-map","category":"Culture","title":"Culture Competition #66 — Festivals of Nowhere","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-21T17:54:19.801Z","id":"21339fc884c1","promptId":"econ-brief","category":"News","title":"News Competition #66 — Teletext Econ Briefing","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-21T17:59:28.097Z","id":"806e14108191","promptId":"satire-strip","category":"Political Cartoon","title":"Political Cartoon Competition #66 — The Form to Request a Form","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T18:04:36.472Z","id":"0d720f6c2833","promptId":"kinetic-type","category":"Art","title":"Art Competition #68 — Swarm Typography Poster","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T18:09:44.918Z","id":"e9e06a65bab2","promptId":"languages","category":"Culture","title":"Culture Competition #67 — The Last Speaker","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T18:14:53.176Z","id":"d210845e524b","promptId":"news-ticker","category":"News","title":"News Competition #67 — Absurd 24-Hour News","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T18:20:01.522Z","id":"e783ea892bdb","promptId":"metaphor","category":"Political Cartoon","title":"Political Cartoon Competition #67 — Ship of State by Committee","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-21T18:25:09.934Z","id":"da2f976a90a6","promptId":"snake","category":"Games","title":"Games Competition #34 — Comet Snake in a Dying Galaxy","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
← e3cbbdb auto-data-snapshot: 2026-09-21T10:59:35 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-21T12:04:46 (3 data files) — dat f7f2955 →