Gemini 3.1 Flash Lite Thinking High - 2,526 tokens:

https://gistpreview.github.io/?3496285c5dac5ba10ebbc0b201a1a...

Gemini 2.5 Pro - 5,325 tokens:

https://gistpreview.github.io/?cc5e0fefeaaffecd228c16c95e736...

Gemini 2.5 Flash - 7,556 tokens:

https://gistpreview.github.io/?263d6058fe526a62b8f270f0620ec...

Gemma 4 31B IT - 3,261 tokens via AI Studio:

https://gistpreview.github.io/?858a42b96af864859a3b89508619d...

Gemma 4 26B A4B IT - 4,034 tokens via AI Studio:

https://gistpreview.github.io/?4adb7703897e0c6b583f9de928e4a...

Gemma 4 E4B it via Edge Gallery on pixel phone:

https://gistpreview.github.io/?da742884e5e830ce71ee4db877519...

OFC this is just for fun, but nevertheless gave me working code on first try.

I'm surprised that, "they must have trained for it" camp is not here saying that rubbish.

Here is GPT 5.5 High thinking; I had to add a second follow up prompt "it's not animated though" as the first one was not animated.

https://gistpreview.github.io/?557f979c82701862bc26d24f10399...

Here is a GPT 5.5 Extra High with a modified instruction:

> Create animated SVG of a frog on a boat rowing through jungle river. Single page self contained HTML page with SVG. Use the Brave Browser to verifty that the image is indeed animated and looks like a proper rowing frog; iterate until you are satisfied with it.

It was able to discover and fix an animation bug, but the result is still far from perfect: https://gistpreview.github.io/?029df86d03bfe8f87df1e4d9ed2f6...

Why is it fixated on the front perspective? Interesting choice though, because most humans (and seems like other LLMs too) would pick a side perspective

Opus 4.7

https://claude.ai/public/artifacts/128ebe5a-add7-406a-9bce-6...

Wow that's terrible. Any idea why?

Did you see the other ones? This is very good by comparison.

Yeah, the oars being around (inverted) is very distracting but the other elements appear quaint and "accurate".

Ah, of course it's all subjective: I was pretty impressed with the Gemini ones. How can the frog move the oars the wrong way around, against each other??

I think Anthropic optimizes less for visuals. Also, it’s not that terrible.

My guess will be because this is just software that don't understand how the world works and it's only trying to please?idk maybe im wrong

These are hilarious. 3.5 Flash Thinking High is the only one that is weirdly deformed (what is going on with the hat in 3.1 Pro??)

hesamation/Qwen3.6-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-GGUF @ Q6_K

8112 tokens @ 52.97 TPS, 0.85s TTFT

https://gistpreview.github.io/?7bdefff99aca89d1bc12405323bd4...

Full session: https://gist.github.com/abtinf/7bdefff99aca89d1bc12405323bd4...

Generated with LM Studio on a Macbook Pro M2 Max

https://huggingface.co/hesamation/Qwen3.6-35B-A3B-Claude-4.6...

It isn’t animated at all for me?

It is animated just no movement like on my 3.5 flash examples. Try different browser might be unless it iOS.

It is animated but the viewer is broken for some reason (tested Chrome latest windows).

This one works:

https://www.svgviewer.dev/s/04ipQgsU

Well, honestly this is quite impressive compared to 3.1 Flash Lite and 2.5 Pro. Considering that 2.5 Pro is actually quite good at generating massive amounts of code one shot.

All three links animate for me.

I think they mean the boat is moving. In the flash ones the paddles are animated but the boat is stationary for me.

The boat moves in all three for me

The boat itself rocks, but do you see the background changing to indicate the boat is progressing through the environment? I only see that in the 3.1 Pro example. I believe that's what the OP meant.

I think this illustrates the problem with OP's prompt. If the goal is specifically to implement a scrolling background, this should have been in the prompt.

Yup. My bad. It was just first idea that come to my mind since I enjoy visually compare each new release with unique prompts.

Can you try with a more complex story such as "three little pigs"? I tried but it created a storybook instead of the SVG animation. I am looking to partially imitate Godogen [1][2] which is really great, even for animations.

[1] https://github.com/htdt/godogen

[2] https://drive.google.com/file/d/1ozZmWcSwieZQG0muYjbj7Xjhhlz...

I think it's unreasonable to expect models generate complex stories in single prompt since they trained to be concise, but I tried. This is prompt on top of story with no control buttons request:

   Now think, plan how to tell this story in a cartoon, make scene outline and then generate SVG animation story for "Three Little Pigs" in self contained HTML page. Just single animation no control buttons.

Full prompt in gist comments: https://gist.github.com/ArseniyShestakov/ed9faa53604035005ca...

Actual results for models, one shot:

Gemini 3.5 Flash - Three Little Pigs - 9,050 tokens:

https://gistpreview.github.io/?ed9faa53604035005cae86c63c766...

Gemini 3.1 Pro - Three Little Pigs - 24,272 tokens:

https://gistpreview.github.io/?f506bbfd9b4459c8cd55d89605af8...

Gemini 3.5 Flash - Three Little Pigs - 9,050 tokens:

https://gistpreview.github.io/?ed9faa53604035005cae86c63c766...

Gemma 4 31B IT - Three Little Pigs - 5,494 tokens:

https://gistpreview.github.io/?a3aa75abbe8fd7818b73f6fa55ee6...

Gemma 4 26B A4B IT - Three Iittle Pigs - 6,375 tokens:

https://gistpreview.github.io/?1e631caebeb54f9f0cd6d0e3d4d5e...

Wow, Gemini 3.5 Flash surprised me there.

3.1 pro was pretty good among them. (iOS)

This was generated locally with Kimi https://gistpreview.github.io/?d55f07c22d54badc8042a7c8b3785...

What Kimi exactly? What version and quant?

K2.6, Qx/Q4, it's huge and mostly runs off CPU/system ram. So slow

I'd say its as good as Gemimi 3.1 Pro for this specific test.

Your links are broken FYI.

They work for me.

They do work here too.

3.5 Flash definitely got the synth wave vibe preference.

It’s shocking how much better 3.1 is than 3.5 flash

The benchmarks used don’t really give a full story