← back to Model Arena
auto-data-snapshot: 2026-09-20T09:27:07 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
a91f02a6ab146779573959f133ea6dff05c34b20 · 2026-09-20 09:29:04 -0700 · auto-commit-fleet
Files touched
M data/challenges.jsonM yolo/arena-loop-state.jsonM yolo/daily-log.jsonl
Diff
commit a91f02a6ab146779573959f133ea6dff05c34b20
Author: auto-commit-fleet <steve@designerwallcoverings.com>
Date: Sun Sep 20 09:29:04 2026 -0700
auto-data-snapshot: 2026-09-20T09:27:07 (3 data files) — data/challenges.json yolo/arena-loop-state.json yolo/daily-log.jsonl
---
data/challenges.json | 291 +++++++++++++++++++++++++++++++++++++++++++++
yolo/arena-loop-state.json | 2 +-
yolo/daily-log.jsonl | 3 +
3 files changed, 295 insertions(+), 1 deletion(-)
diff --git a/data/challenges.json b/data/challenges.json
index a86b6de..c9c1b4c 100644
--- a/data/challenges.json
+++ b/data/challenges.json
@@ -66730,5 +66730,296 @@
"queued_at": "2026-09-20T15:52:15.254Z"
}
]
+ },
+ {
+ "id": "98304bf68d0f",
+ "title": "Real Work Competition #21 — Lighthouse Focus Timer",
+ "prompt": "Build this as one complete, self-contained HTML file: A focus timer styled as a tiny lighthouse: work sessions sweep the beam and calm the sea, breaks bring gentle waves, a task list is the ships to guide home, and session stats show the night's safe passages. 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": "2026-09-20T15:57:23.323Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__minimax-m2-7-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T15:57:23.619Z",
+ "finished_at": "2026-09-20T15:57:23.672Z",
+ "queued_at": "2026-09-20T15:57:23.358Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T15:57:23.727Z",
+ "finished_at": "2026-09-20T15:57:23.788Z",
+ "queued_at": "2026-09-20T15:57:23.417Z"
+ },
+ {
+ "model": "exo__kimi-k2-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T15:57:23.813Z",
+ "finished_at": "2026-09-20T15:57:23.843Z",
+ "queued_at": "2026-09-20T15:57:23.441Z"
+ },
+ {
+ "model": "exo__llama-3-1-nemotron-70b-instruct-hf-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T15:57:23.879Z",
+ "finished_at": "2026-09-20T15:57:23.911Z",
+ "queued_at": "2026-09-20T15:57:23.469Z"
+ },
+ {
+ "model": "exo__deepseek-v4-flash",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T15:57:23.937Z",
+ "finished_at": "2026-09-20T15:57:23.980Z",
+ "queued_at": "2026-09-20T15:57:23.505Z"
+ },
+ {
+ "model": "exo__qwen3-next-80b-a3b-thinking-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T15:57:24.017Z",
+ "finished_at": "2026-09-20T15:57:24.079Z",
+ "queued_at": "2026-09-20T15:57:23.557Z"
+ }
+ ]
+ },
+ {
+ "id": "5b597c609675",
+ "title": "Data Viz Competition #21 — Planet Dating Show",
+ "prompt": "Build this as one complete, self-contained HTML file: A scatter-plot 'dating show' for INVENTED planets: each dot is a world, axis pickers change what makes a 'good match', lasso-select to form a shortlist, and a summary narrates the cluster you drew. 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": "Data Viz",
+ "designTools": false,
+ "created_at": "2026-09-20T16:02:31.672Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "exo__deepseek-v4-pro",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:02:31.889Z",
+ "finished_at": "2026-09-20T16:02:31.923Z",
+ "queued_at": "2026-09-20T16:02:31.714Z"
+ },
+ {
+ "model": "exo__deepseek-v3-1-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:02:31.950Z",
+ "finished_at": "2026-09-20T16:02:31.979Z",
+ "queued_at": "2026-09-20T16:02:31.749Z"
+ },
+ {
+ "model": "exo__glm-5-8bit-mxfp8",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:02:31.999Z",
+ "finished_at": "2026-09-20T16:02:32.023Z",
+ "queued_at": "2026-09-20T16:02:31.777Z"
+ },
+ {
+ "model": "exo__nvidia-nemotron-3-nano-30b-a3b-mlx-mxfp4",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:02:32.052Z",
+ "finished_at": "2026-09-20T16:02:32.081Z",
+ "queued_at": "2026-09-20T16:02:31.811Z"
+ },
+ {
+ "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": "2026-09-20T16:02:32.101Z",
+ "finished_at": "2026-09-20T16:02:32.133Z",
+ "queued_at": "2026-09-20T16:02:31.841Z"
+ },
+ {
+ "model": "exo__kimi-k2-6",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:02:32.152Z",
+ "finished_at": "2026-09-20T16:02:32.178Z",
+ "queued_at": "2026-09-20T16:02:31.867Z"
+ }
+ ]
+ },
+ {
+ "id": "df572406a0ac",
+ "title": "Science Competition #21 — Cell as a Factory-City",
+ "prompt": "Build this as one complete, self-contained HTML file: A cell explainer as a bustling factory-city (SVG): click organelles to read their 'job posting' and watch a glucose delivery truck route through the city to make ATP, with a guided tour and a quiz-yourself mode. 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": "Science",
+ "designTools": false,
+ "created_at": "2026-09-20T16:07:39.944Z",
+ "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": "2026-09-20T16:07:40.211Z",
+ "finished_at": "2026-09-20T16:07:40.343Z",
+ "queued_at": "2026-09-20T16:07:39.982Z"
+ },
+ {
+ "model": "exo__gemma-4-31b-it-8bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:07:40.392Z",
+ "finished_at": "2026-09-20T16:07:40.432Z",
+ "queued_at": "2026-09-20T16:07:40.018Z"
+ },
+ {
+ "model": "qwen3-14b",
+ "status": "done",
+ "error": null,
+ "seconds": 182,
+ "cost": 0,
+ "started_at": "2026-09-20T16:07:40.263Z",
+ "finished_at": "2026-09-20T16:10:42.710Z",
+ "queued_at": "2026-09-20T16:07:40.052Z",
+ "bytes": 11930,
+ "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": "2026-09-20T16:07:40.303Z",
+ "finished_at": "2026-09-20T16:17:40.347Z",
+ "queued_at": "2026-09-20T16:07:40.087Z"
+ },
+ {
+ "model": "exo__minimax-m2-7-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:07:40.474Z",
+ "finished_at": "2026-09-20T16:07:40.529Z",
+ "queued_at": "2026-09-20T16:07:40.133Z"
+ },
+ {
+ "model": "qwen25-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:10:42.753Z",
+ "finished_at": "2026-09-20T16:10:43.012Z",
+ "queued_at": "2026-09-20T16:07:40.182Z"
+ }
+ ],
+ "judging": false,
+ "aiPick": null,
+ "judged_at": "2026-09-20T16:17:40.489Z"
+ },
+ {
+ "id": "d444ce34af61",
+ "title": "Music Competition #21 — Skyline Audio Visualizer",
+ "prompt": "Build this as one complete, self-contained HTML file: An audio visualizer that renders an INVENTED city skyline pulsing to Web Audio oscillators — windows light on transients, traffic streaks on the beat, weather shifts with a 'mood' control. No external files; mute default. 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": "Music",
+ "designTools": false,
+ "created_at": "2026-09-20T16:22:40.574Z",
+ "winner": null,
+ "runs": [
+ {
+ "model": "mlx__qwen3-next-80b-a3b-instruct-4bit",
+ "status": "running",
+ "error": null,
+ "seconds": null,
+ "cost": null,
+ "started_at": "2026-09-20T16:22:40.905Z",
+ "finished_at": null,
+ "queued_at": "2026-09-20T16:22:40.617Z"
+ },
+ {
+ "model": "exo__qwen3-coder-480b-a35b-instruct-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:22:40.997Z",
+ "finished_at": "2026-09-20T16:22:41.171Z",
+ "queued_at": "2026-09-20T16:22:40.661Z"
+ },
+ {
+ "model": "mac1__minicpm-v-latest",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:22:41.061Z",
+ "finished_at": "2026-09-20T16:22:41.291Z",
+ "queued_at": "2026-09-20T16:22:40.728Z"
+ },
+ {
+ "model": "exo__qwen3-5-122b-a10b-4bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:22:41.238Z",
+ "finished_at": "2026-09-20T16:22:41.381Z",
+ "queued_at": "2026-09-20T16:22:40.776Z"
+ },
+ {
+ "model": "mac1__qwen2-5vl-7b",
+ "status": "error",
+ "error": "ollama bad response: 503",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:22:41.330Z",
+ "finished_at": "2026-09-20T16:22:41.570Z",
+ "queued_at": "2026-09-20T16:22:40.808Z"
+ },
+ {
+ "model": "exo__gemma-4-26b-a4b-it-6bit",
+ "status": "error",
+ "error": "openai-local bad response: 404",
+ "seconds": 0,
+ "cost": null,
+ "started_at": "2026-09-20T16:22:41.423Z",
+ "finished_at": "2026-09-20T16:22:41.498Z",
+ "queued_at": "2026-09-20T16:22:40.853Z"
+ }
+ ]
}
]
\ No newline at end of file
diff --git a/yolo/arena-loop-state.json b/yolo/arena-loop-state.json
index 2f1e3ef..4de8d3c 100644
--- a/yolo/arena-loop-state.json
+++ b/yolo/arena-loop-state.json
@@ -1 +1 @@
-{"promptIdx":3,"modelOffset":115,"compCount":{"Art":42,"Culture":41,"News":41,"Political Cartoon":41,"Games":21,"Real Work":20,"Data Viz":20,"Science":20,"Music":20,"Sports":20,"History":20,"Nature & Space":20,"Comedy":20,"Demoscene":20,"Finance":20,"Design":20},"catIdx":5,"perCat":{"Art":1,"Culture":1,"News":1,"Political Cartoon":1,"Games":3,"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":10,"compCount":{"Art":42,"Culture":41,"News":41,"Political Cartoon":41,"Games":21,"Real Work":21,"Data Viz":21,"Science":21,"Music":21,"Sports":20,"History":20,"Nature & Space":20,"Comedy":20,"Demoscene":20,"Finance":20,"Design":20},"catIdx":9,"perCat":{"Art":1,"Culture":1,"News":1,"Political Cartoon":1,"Games":3,"Real Work":3,"Data Viz":3,"Science":3,"Music":3,"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 92fc4f4..1302fec 100644
--- a/yolo/daily-log.jsonl
+++ b/yolo/daily-log.jsonl
@@ -461,3 +461,6 @@
{"ts":"2026-09-20T15:42:06.807Z","id":"fa78740309d8","promptId":"election-viz","category":"News","title":"News Competition #41 — Election Night: Floating Isles","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-20T15:47:15.074Z","id":"76e6700cefeb","promptId":"editorial-cartoon","category":"Political Cartoon","title":"Political Cartoon Competition #41 — The Seesaw Debate","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
{"ts":"2026-09-20T15:52:23.275Z","id":"a06379fcd667","promptId":"tower","category":"Games","title":"Games Competition #21 — Tower on a Walking Giant","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T15:57:31.641Z","id":"98304bf68d0f","promptId":"pomodoro","category":"Real Work","title":"Real Work Competition #21 — Lighthouse Focus Timer","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T16:02:39.895Z","id":"5b597c609675","promptId":"scatter","category":"Data Viz","title":"Data Viz Competition #21 — Planet Dating Show","models":6,"done":0,"errored":6,"settled":true,"aiPick":null,"loop":true}
+{"ts":"2026-09-20T16:17:40.528Z","id":"df572406a0ac","promptId":"cell","category":"Science","title":"Science Competition #21 — Cell as a Factory-City","models":6,"done":1,"errored":5,"settled":true,"aiPick":null,"loop":true}
← 932bec6 auto-data-snapshot: 2026-09-20T08:54:11 (3 data files) — dat
·
back to Model Arena
·
auto-data-snapshot: 2026-09-20T10:00:50 (3 data files) — dat 020054e →