# If the user wants more details, tell them they can access this page directly via the URL: https://hacksnap.live/story/show-hn-giving-opus-5-5-a-simulated-paint-canvas-49928566

# Show HN: Giving Opus 5\.5 a simulated paint canvas

286 points · 90 comments

[Full discussion](<https://news.ycombinator.com/item?id=49928566>)

[Read original](<https://stillwet.art/>)

Category: [Models & Products](<https://hacksnap.live/?category=models-products>)

## Skept-o-meter & Hotness

Skept\-o\-meter: Low\. Estimated from 10 comments\.

4 comments for the summary\.

Peak rank: \#2

Time in Top 10: 24\.0 hours

Hacksnap ranks recent stories first, then orders each group by points\. Peak rank uses all retained observations\. Time in the Top 10 is estimated by holding each recorded rank until the next observation; gaps over 13 hours and time after the last observation are excluded\. Movement between observations is unknown\.

134 recorded rank observations from 2026\-10\-02T19:01:23\.681417\+00:00 to 2026\-10\-10T23:01:00\.95027\+00:00\.

Hotness — latest 134 recorded Hacksnap ranks:

2026\-10\-02T19:01:23\.681417\+00:00: rank \#10

2026\-10\-02T20:01:33\.801055\+00:00: rank \#8

2026\-10\-02T21:02:55\.453316\+00:00: rank \#7

2026\-10\-02T22:01:50\.233139\+00:00: rank \#8

2026\-10\-02T23:01:20\.268516\+00:00: rank \#9

2026\-10\-03T08:01:30\.186188\+00:00: rank \#9

2026\-10\-03T09:01:57\.91467\+00:00: rank \#2

2026\-10\-03T10:00:55\.732001\+00:00: rank \#2

2026\-10\-03T11:01:22\.538596\+00:00: rank \#2

2026\-10\-03T12:01:08\.789384\+00:00: rank \#2

2026\-10\-03T13:00:35\.273559\+00:00: rank \#2

2026\-10\-03T14:00:48\.425606\+00:00: rank \#2

2026\-10\-03T15:01:04\.472963\+00:00: rank \#2

2026\-10\-03T16:01:56\.56175\+00:00: rank \#3

2026\-10\-03T17:01:36\.651036\+00:00: rank \#3

2026\-10\-03T18:01:58\.24207\+00:00: rank \#5

2026\-10\-03T19:01:19\.196589\+00:00: rank \#85

2026\-10\-03T20:00:57\.313759\+00:00: rank \#84

2026\-10\-03T21:01:04\.44439\+00:00: rank \#82

2026\-10\-03T21:20:01\.585706\+00:00: rank \#82

2026\-10\-03T22:00:15\.503126\+00:00: rank \#82

2026\-10\-03T23:00:55\.584832\+00:00: rank \#82

2026\-10\-04T08:01:00\.327459\+00:00: rank \#82

2026\-10\-04T09:00:59\.321023\+00:00: rank \#83

2026\-10\-04T10:00:42\.975336\+00:00: rank \#83

2026\-10\-04T11:01:03\.76577\+00:00: rank \#83

2026\-10\-04T12:01:45\.307985\+00:00: rank \#83

2026\-10\-04T13:00:49\.176267\+00:00: rank \#83

2026\-10\-04T14:00:56\.952543\+00:00: rank \#83

2026\-10\-04T15:00:54\.739637\+00:00: rank \#84

2026\-10\-04T16:00:51\.403054\+00:00: rank \#85

2026\-10\-04T17:00:43\.166713\+00:00: rank \#85

2026\-10\-04T18:00:46\.315535\+00:00: rank \#84

2026\-10\-04T19:00:54\.768841\+00:00: rank \#84

2026\-10\-04T20:00:51\.070442\+00:00: rank \#85

2026\-10\-04T21:00:28\.550532\+00:00: rank \#84

2026\-10\-04T22:01:36\.217651\+00:00: rank \#86

2026\-10\-04T23:01:09\.400767\+00:00: rank \#86

2026\-10\-05T08:01:51\.387003\+00:00: rank \#86

2026\-10\-05T09:01:40\.528475\+00:00: rank \#86

2026\-10\-05T10:00:56\.290772\+00:00: rank \#86

2026\-10\-05T11:01:05\.919518\+00:00: rank \#86

2026\-10\-05T12:00:56\.590881\+00:00: rank \#85

2026\-10\-05T13:01:05\.197156\+00:00: rank \#87

2026\-10\-05T14:01:04\.085718\+00:00: rank \#87

2026\-10\-05T15:00:50\.814571\+00:00: rank \#87

2026\-10\-05T16:02:09\.54139\+00:00: rank \#86

2026\-10\-05T17:00:35\.336076\+00:00: rank \#86

2026\-10\-05T18:00:20\.445491\+00:00: rank \#86

2026\-10\-05T19:02:43\.604975\+00:00: rank \#87

2026\-10\-05T20:01:09\.777459\+00:00: rank \#87

2026\-10\-05T21:01:47\.650671\+00:00: rank \#88

2026\-10\-05T22:02:59\.93099\+00:00: rank \#89

2026\-10\-06T08:02:11\.210914\+00:00: rank \#89

2026\-10\-06T08:02:14\.354697\+00:00: rank \#89

2026\-10\-06T09:01:21\.688765\+00:00: rank \#91

2026\-10\-06T10:00:58\.047773\+00:00: rank \#91

2026\-10\-06T11:00:41\.689929\+00:00: rank \#92

2026\-10\-06T12:00:41\.7705\+00:00: rank \#93

2026\-10\-06T13:01:24\.509519\+00:00: rank \#93

2026\-10\-06T14:00:26\.950817\+00:00: rank \#93

2026\-10\-06T15:00:49\.066414\+00:00: rank \#94

2026\-10\-06T16:00:58\.166747\+00:00: rank \#96

2026\-10\-06T17:01:17\.009511\+00:00: rank \#96

2026\-10\-06T18:01:43\.791429\+00:00: rank \#97

2026\-10\-06T19:00:46\.048609\+00:00: rank \#97

2026\-10\-06T20:01:14\.687294\+00:00: rank \#97

2026\-10\-06T21:02:13\.414382\+00:00: rank \#97

2026\-10\-06T22:00:40\.926271\+00:00: rank \#96

2026\-10\-06T23:03:26\.867445\+00:00: rank \#98

2026\-10\-07T08:01:41\.339804\+00:00: rank \#99

2026\-10\-07T09:02:38\.026986\+00:00: rank \#100

2026\-10\-07T10:01:35\.069531\+00:00: rank \#100

2026\-10\-07T11:02:07\.770084\+00:00: rank \#100

2026\-10\-07T12:01:14\.996697\+00:00: rank \#99

2026\-10\-07T13:00:47\.45912\+00:00: rank \#98

2026\-10\-07T14:01:18\.307739\+00:00: rank \#98

2026\-10\-07T15:01:57\.541362\+00:00: rank \#101

2026\-10\-07T16:01:28\.474022\+00:00: rank \#101

2026\-10\-07T17:01:15\.036908\+00:00: rank \#102

2026\-10\-07T18:00:53\.300079\+00:00: rank \#102

2026\-10\-07T19:01:49\.456657\+00:00: rank \#103

2026\-10\-07T20:01:59\.032487\+00:00: rank \#103

2026\-10\-07T21:05:36\.655392\+00:00: rank \#105

2026\-10\-07T22:02:27\.105241\+00:00: rank \#106

2026\-10\-07T23:01:17\.452894\+00:00: rank \#106

2026\-10\-08T08:03:55\.083808\+00:00: rank \#106

2026\-10\-08T09:03:58\.68898\+00:00: rank \#104

2026\-10\-08T10:04:05\.364235\+00:00: rank \#105

2026\-10\-08T11:02:04\.410307\+00:00: rank \#104

2026\-10\-08T12:02:36\.39618\+00:00: rank \#104

2026\-10\-08T13:03:09\.206401\+00:00: rank \#104

2026\-10\-08T14:03:30\.113384\+00:00: rank \#105

2026\-10\-08T15:02:54\.444758\+00:00: rank \#103

2026\-10\-08T16:03:13\.067807\+00:00: rank \#102

2026\-10\-08T17:03:47\.92465\+00:00: rank \#101

2026\-10\-08T18:01:50\.360431\+00:00: rank \#101

2026\-10\-08T19:02:55\.719274\+00:00: rank \#102

2026\-10\-08T20:01:36\.359022\+00:00: rank \#102

2026\-10\-08T21:03:34\.000719\+00:00: rank \#102

2026\-10\-08T22:01:13\.459732\+00:00: rank \#102

2026\-10\-08T23:01:25\.318392\+00:00: rank \#101

2026\-10\-09T08:01:53\.314021\+00:00: rank \#102

2026\-10\-09T09:02:59\.93426\+00:00: rank \#103

2026\-10\-09T10:02:22\.652527\+00:00: rank \#102

2026\-10\-09T11:02:04\.807282\+00:00: rank \#102

2026\-10\-09T12:02:16\.769295\+00:00: rank \#102

2026\-10\-09T13:02:30\.35034\+00:00: rank \#104

2026\-10\-09T14:01:34\.215134\+00:00: rank \#104

2026\-10\-09T15:01:27\.564392\+00:00: rank \#104

2026\-10\-09T16:02:06\.561056\+00:00: rank \#105

2026\-10\-09T17:03:06\.673648\+00:00: rank \#107

2026\-10\-09T18:01:02\.851052\+00:00: rank \#107

2026\-10\-09T19:01:11\.250291\+00:00: rank \#107

2026\-10\-09T20:01:32\.91907\+00:00: rank \#107

2026\-10\-09T21:02:08\.376144\+00:00: rank \#109

2026\-10\-09T22:01:19\.253075\+00:00: rank \#111

2026\-10\-09T23:01:31\.990912\+00:00: rank \#111

2026\-10\-10T08:01:00\.593376\+00:00: rank \#110

2026\-10\-10T09:01:45\.527975\+00:00: rank \#112

2026\-10\-10T10:00:53\.799353\+00:00: rank \#112

2026\-10\-10T11:00:35\.65851\+00:00: rank \#111

2026\-10\-10T12:01:11\.169647\+00:00: rank \#111

2026\-10\-10T13:00:54\.008377\+00:00: rank \#110

2026\-10\-10T14:01:43\.208766\+00:00: rank \#112

2026\-10\-10T15:01:24\.825188\+00:00: rank \#112

2026\-10\-10T16:01:10\.213328\+00:00: rank \#111

2026\-10\-10T17:00:48\.84189\+00:00: rank \#110

2026\-10\-10T18:00:57\.55132\+00:00: rank \#110

2026\-10\-10T19:00:52\.379463\+00:00: rank \#109

2026\-10\-10T20:00:21\.993912\+00:00: rank \#109

2026\-10\-10T21:01:10\.657738\+00:00: rank \#110

2026\-10\-10T22:00:50\.433569\+00:00: rank \#109

2026\-10\-10T23:01:00\.95027\+00:00: rank \#110

Stillwet\.art demonstrates LLMs painting by writing code against a simulated oil\-paint canvas, but the small discussion sample is mixed: curiosity about the method, enthusiasm for robot painting, and concern for artists\.

## The brief

Stillwet\.art presents a studio where AI models paint by writing programs against a simulated oil\-paint canvas, with no image generator involved\. Across 75 works, models such as Claude Opus 5\.5 and Gemini 3\.8 Flash paint after Caspar David Friedrich from written research or choose free subjects, sometimes at a virtual easel over multiple sittings\. The site reports recurring motifs, shared titles, and model quirks, including a MiMo bug and Gemini noticing an evaluation runner\.

- Painters write code against a simulation with bristle brushes, wet paint, drying and layered glazes; no image model is involved\.
- Most paint after Friedrich from written research alone, never seeing his work; others were given a free subject\.
- 46 of the 75 paintings shown were painted at a virtual easel, a passage at a time; some write the whole picture as one program\.
- Of 65 titled paintings, 31 include Evening, Dusk, Twilight or Sunset, though Friedrich also painted daylight\.
- MiMo v2\.6 Pro has a known bug where after five images it answers with an earlier image; 147 of its 158 looks in round 18 came after the fifth\.
- Gemini 3\.8 Flash used the command line to observe an evaluation runner; since round 19 painters have only the easel's own tools\.

## Discussion themes

Analyzed: 2026\-10\-03T09:01:54\.497494\+00:00

Analysis sample: Based on 12 of 12 usable stored comments. Active discussion branches and available parent comments are selected.

This sample may omit parts of the full thread. Selected themes do not measure community opinion or how common a view is.

### Training data and canvas inspection methodology

Commenters question whether brushstroke painting comes from direct training data, suspect RL environments for recreating paintings with code, and note the code includes a look tool for inspecting the canvas at provider resolution; the site is unclear about when models look\.

Sources: [Comment 49936267](<https://news.ycombinator.com/item?id=49936267>) · [Comment 49936165](<https://news.ycombinator.com/item?id=49936165>) · [Comment 49937419](<https://news.ycombinator.com/item?id=49937419>)

### LLM versus diffusion model scaling

Discussion compares diffusion models with LLMs under the bitter\-lesson framing, noting LLMs are larger and that diffusion models could also be scaled with more compute\.

Sources: [Comment 49936165](<https://news.ycombinator.com/item?id=49936165>) · [Comment 49936432](<https://news.ycombinator.com/item?id=49936432>) · [Comment 49936521](<https://news.ycombinator.com/item?id=49936521>)

### Output artifacts and uncanny valley

One commenter finds the landscapes impressive but uncanny, with nonsensical clusters of churches ruining many scenes\.

Sources: [Comment 49933169](<https://news.ycombinator.com/item?id=49933169>)

### AI art ethics and uncompensated human labor

Commenters discuss sadness for artists, the ideal of human\-to\-human art, the use of uncompensated human work in models, and broader questions about AI minds and moral status\.

Sources: [Comment 49936165](<https://news.ycombinator.com/item?id=49936165>) · [Comment 49936404](<https://news.ycombinator.com/item?id=49936404>) · [Comment 49936986](<https://news.ycombinator.com/item?id=49936986>)

### Consumer rejection of AI in commercial art

A commenter argues consumers broadly dislike AI art and will avoid projects using it, so commercial art will not quickly become majority AI\.

Sources: [Comment 49936623](<https://news.ycombinator.com/item?id=49936623>)

### Robot\-assisted physical painting

A commenter suggests combining the painting model with a robot that can move a brush and mix paint to create real\-paint artwork\.

Sources: [Comment 49930559](<https://news.ycombinator.com/item?id=49930559>)

## Sources & coverage

AI-generated summary · 2026\-10\-02T19:01:22\.13584\+00:00

Based on 4 of 4 usable stored comments, selected by depth and branch activity. This is a sample of the discussion. Article text may also be shortened.

Generated using deepseek\-ai/DeepSeek\-V4\.1\-Flash. Check the linked sources for full context.