PHASE 01 · Completed

Phase 1 · 15 benchmark tasks

SVG illustration, single-file interactive apps, and multi-file web applications.

Tasks planned for this phase (Task 01–15), per PROMPT/Phase1_TEST_PROMPTS_BILINGUAL.md.
0115 / 15 Tasks4 Models60 Runs
Archived task versionstask-01@1 · task-02@1 · task-03@1 · task-04@1 · task-05@1 · task-06@1 · task-07@1 · task-08@1 · task-09@1 · task-10@1 · task-11@1 · task-12@1 · task-13@1 · task-14@1 · task-15@1
Total duration7:16:43
ModelsDeepSeek-V4.1-Flash (0910 preview + GA release) / Muse Spark 1.3 / GPT-5.6 Sol / K3
SELECTED WORKS

Selected works

View all tasks →
PER-TASK RECORDS

Per-task records

01

Aevum Luxury Watch Landing Page

website

DeepSeek-V4.1-Flash (0910 preview + GA release)
4:51 · 24 API calls
Muse Spark 1.3
2:13 · 10 API calls
GPT-5.6 Sol
6:06 · 4 API calls
K3
33:08 · 21 API calls
View task →
02

Pelican Riding a Bicycle

svg-illustration

DeepSeek-V4.1-Flash (0910 preview + GA release)
14:38 · 48 API calls
Muse Spark 1.3
0:50 · 5 API calls
GPT-5.6 Sol
2:06 · 3 API calls
K3
6:55 · 11 API calls
View task →
03

Analog Clock at 6:25

svg-clock

DeepSeek-V4.1-Flash (0910 preview + GA release)
0:51 · 6 API calls
Muse Spark 1.3
0:39 · 8 API calls
GPT-5.6 Sol
2:15 · 8 API calls
K3
2:16 · 5 API calls
View task →
04

Analog Clock at 11:52:30

svg-clock

DeepSeek-V4.1-Flash (0910 preview + GA release)
0:18 · 2 API calls
Muse Spark 1.3
0:36 · 6 API calls
GPT-5.6 Sol
2:02 · 3 API calls
K3
4:29 · 5 API calls
View task →
05

Side-View Bicycle Rider

svg-illustration

DeepSeek-V4.1-Flash (0910 preview + GA release)
1:20 · 6 API calls
Muse Spark 1.3
2:51 · 18 API calls
GPT-5.6 Sol
2:19 · 4 API calls
K3
11:56 · 18 API calls
View task →
06

Hand Cutting Paper with Scissors

svg-illustration

DeepSeek-V4.1-Flash (0910 preview + GA release)
1:11 · 3 API calls
Muse Spark 1.3
2:07 · 15 API calls
GPT-5.6 Sol
2:22 · 4 API calls
K3
11:54 · 5 API calls
View task →
07

Person Pushing a Wheelbarrow

svg-illustration

DeepSeek-V4.1-Flash (0910 preview + GA release)
0:48 · 3 API calls
Muse Spark 1.3
0:40 · 7 API calls
GPT-5.6 Sol
3:09 · 12 API calls
K3
4:19 · 5 API calls
View task →
08

Single-File Breakout Game

game

DeepSeek-V4.1-Flash (0910 preview + GA release)
11:57 · 46 API calls
Muse Spark 1.3
1:18 · 6 API calls
GPT-5.6 Sol
7:03 · 13 API calls
K3
2:53 · 5 API calls
View task →
09

Single-File Top-Down Farming Game

game

DeepSeek-V4.1-Flash (0910 preview + GA release)
11:34 · 42 API calls
Muse Spark 1.3
1:41 · 7 API calls
GPT-5.6 Sol
5:10 · 4 API calls
K3
11:01 · 15 API calls
View task →
10

Pixel-Art Editor

creative-tool

DeepSeek-V4.1-Flash (0910 preview + GA release)
7:17 · 34 API calls
Muse Spark 1.3
0:46 · 7 API calls
GPT-5.6 Sol
6:14 · 7 API calls
K3
6:23 · 9 API calls
View task →
11

2D Floor-Plan Editor

productivity-tool

DeepSeek-V4.1-Flash (0910 preview + GA release)
12:10 · 81 API calls
Muse Spark 1.3
1:43 · 12 API calls
GPT-5.6 Sol
9:04 · 22 API calls
K3
18:54 · 19 API calls
View task →
12

Calculator App

utility

DeepSeek-V4.1-Flash (0910 preview + GA release)
6:20 · 89 API calls
Muse Spark 1.3
2:00 · 18 API calls
GPT-5.6 Sol
6:04 · 16 API calls
K3
4:05 · 12 API calls
View task →
13

Weather Dashboard

data-dashboard

DeepSeek-V4.1-Flash (0910 preview + GA release)
27:17 · 111 API calls
Muse Spark 1.3
4:46 · 30 API calls
GPT-5.6 Sol
6:29 · 6 API calls
K3
51:23 · 65 API calls
View task →
14

Bank Website

website

DeepSeek-V4.1-Flash (0910 preview + GA release)
23:04 · 123 API calls
Muse Spark 1.3
2:00 · 13 API calls
GPT-5.6 Sol
9:31 · 6 API calls
K3
17:28 · 27 API calls
View task →
15

Book-Tracking App

productivity-tool

DeepSeek-V4.1-Flash (0910 preview + GA release)
14:26 · 75 API calls
Muse Spark 1.3
1:35 · 19 API calls
GPT-5.6 Sol
6:52 · 8 API calls
K3
9:06 · 28 API calls
View task →