Which Ollama Cloud Model is Best? Elephant on a Skateboard SVG Comparison (13 Models)

After testing LLMs on ducks, vehicles, dev scenes, marine life, chess, and the FIFA World Cup, we wanted to know: can today’s top models draw a whimsical, physically impossible scene? This time we asked 13 Ollama cloud models to draw an elephant on a skateboard – a prompt that tests organic anatomy (a massive elephant), mechanical precision (a skateboard with wheels), scale mismatch (a huge animal on a tiny board), and creative humor.

The prompt was: Make an svg of elephant on a skateboard

This is the ninth in our SVG benchmark series. See also: duck on a bicycle, duck with a parachute, duck driving a jeep, cherry blossom trees, duck programmer debugging at 3am, baby shark fish, octopus playing chess, and FIFA World Cup 2026.

Why an elephant on a skateboard? This prompt is a delightfully absurd stress test for SVG generation because it combines multiple hard problems: (1) Organic anatomy – an elephant has a massive body, four thick legs, a trunk, large ears (African) or small ears (Asian), tusks, a tail, and wrinkles, all requiring complex <path> curves, (2) Mechanical precision – a skateboard has a specific deck shape, 4 wheels, and 2 trucks (axle assemblies) that must be positioned symmetrically, (3) Scale mismatch – an elephant weighs 6 tons and a skateboard is meant for a 80kg human; the model must decide how to reconcile this (tiny skateboard under a huge elephant? oversized skateboard?), (4) Balance and pose – a skateboarding elephant must somehow balance on the board, which requires the model to think about physics and pose, (5) Concept combination – the model must understand both “elephant” and “skateboard” and combine them in a single coherent scene, (6) Humor and whimsy – the prompt is inherently funny, and a good model should capture that playfulness (e.g., the elephant looking excited, wearing a helmet, or doing a trick). A model that draws a great trophy may fail here because elephants require organic curves, not geometric shapes.

The goal is not to declare a winner – it is to give you the data so you can pick the best model for your own use case. We show you the SVG, the stats, and a short analysis for each. You decide.

How to Choose the Best Ollama Model for Whimsical Animal SVGs

The elephant-skateboard prompt rewards different things than previous prompts. Here are the criteria to use:

• Elephant anatomy: Does the SVG have a trunk, four legs, large ears, a tail, and tusks? Or is it a generic blob? The trunk is the hardest part – it requires complex bezier curves and must be positioned correctly on the face.
• Skateboard accuracy: Does the SVG have a deck (the board), 4 wheels, and 2 trucks? Or is it just a rectangle? A good skateboard has the iconic shape with upturned ends.
• Scale and proportion: How does the model handle the size mismatch? A tiny skateboard under a huge elephant is funny but unrealistic; an oversized skateboard looks more plausible. The model’s choice reveals its creative judgment.
• Pose and balance: Is the elephant standing on the skateboard? Doing a trick? Looking excited? A good scene captures the dynamism of skateboarding, not just a static elephant on a static board.
• Scene context: Did the model add motion lines, a ground/surface, a sky, or other context? Or is the elephant-skateboard floating in a void?
• Humor and whimsy: Does the SVG capture the playfulness of the prompt? Look for details like a helmet, sunglasses, a grin, or a trick pose that show the model “got the joke”.
• SVG code quality: Does it use <defs>, <use>, gradients, and clean structure? Better code is easier to tweak (e.g., to recolor the elephant or change the skateboard design).

How It Works

The script discovers all cloud-hosted models via the Ollama API (/api/tags), pulls each model, then sends the identical prompt through the OpenAI-compatible endpoint (http://localhost:11434/v1/chat/completions). Each model’s response is parsed for an <svg>...</svg> block, and the extracted SVG is saved for rendering with minimal post-processing (adding width="100%" height="auto" for responsive embedding and fixing XML errors so the SVG renders in browsers).

Cloud models are identified by the remote_host field in the API response – these models are hosted on Ollama Cloud rather than running locally. This means even very large models (671B parameters) can be queried instantly without local GPU resources.

Summary Table: Compare All Models at a Glance

Use this table to quickly compare models on the metrics that matter. The verdict column is a one-line summary to help you shortlist – but read the per-model sections below for the full picture before you decide.

#	Model	SVG Size	Shapes	Colors	Complexity	Verdict
1	deepseek-v4-flash:cloud	5554	49	17	High	Detailed
2	deepseek-v4-pro:cloud	8153	67	10	Very high	Richest scene
3	gemma4:31b-cloud	1240	13	6	Compact	Compact
4	gemma4:cloud	1155	12	6	Compact	Compact
5	glm-5.1:cloud	11862	78	26	Very high	Richest scene
6	glm-5.2:cloud	5364	60	18	Very high	Richest scene
7	gpt-oss:120b-cloud	1659	20	7	Medium	Balanced
8	kimi-k2.6:cloud	7511	30	15	High	Detailed
9	minimax-m2.7:cloud	3010	24	8	Medium	Balanced
10	minimax-m3:cloud	5627	63	23	Very high	Richest scene
11	nemotron-3-super:cloud	1995	22	7	Medium	Balanced
12	nemotron-3-ultra:cloud	7239	39	14	High	Detailed
13	qwen3.5:397b-cloud	2604	22	15	Medium	Balanced
14	bjoernb/claude-opus-4-5:latest	-	-	-	-	Retired (410)
15	deepseek-v3.1:671b-cloud	-	-	-	-	Retired (410)
16	glm-5:cloud	-	-	-	-	Retired (410)
17	qwen3-vl:235b-cloud	-	-	-	-	Retired (410)

13 out of 17 active models produced a valid SVG. The 4 retired models returned HTTP 410 Gone (removed from Ollama Cloud on 2026-07-15).

Quick Recommendation by Use Case

If you just want a shortcut, here is which model to pick based on what you care about:

• You want the most detailed elephant skateboard SVG: pick models labeled “Very high” complexity in the table above
• You want the fastest response: look at the per-model sections below for the elapsed time
• You want the cleanest, most reusable SVG code: pick models that use <defs>, <use>, and transforms (see raw source below each SVG)
• You want a small, efficient SVG for web embedding: pick models with “Compact” verdict
• You want accurate elephant anatomy (trunk, ears, tusks): check the per-model analysis – the trunk is the hardest part
• You want a funny/whimsical scene: look for models that added details like motion lines, a helmet, or a trick pose
• You want a balance of detail and speed: pick models labeled “Balanced” or “Detailed”
• You want to compare within a model family: pick deepseek-v4-pro vs deepseek-v4-flash, or glm-5.1 vs glm-5.2, or minimax-m2.7 vs minimax-m3

Now read on for the full per-model breakdown and judge for yourself.

1. deepseek-v4-flash:cloud

SVG size: 5554 characters
Complexity: High
Shape elements: 49
Distinct colors: 17
Raw response: 6112 characters
Generation time: 48.6s

Analysis

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Uses transforms (translate, rotate, scale) for positioning the elephant and tilting the skateboard
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders tusks – the model captured fine anatomical detail
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Renders trucks (the axle assemblies) – the model captured fine skateboard mechanical detail
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface

With 5,554 characters of SVG markup, this is a detailed output that balances richness with readability.

View raw SVG source

```xml ```

2. deepseek-v4-pro:cloud

SVG size: 8153 characters
Complexity: Very high
Shape elements: 67
Distinct colors: 10
Raw response: 8564 characters
Generation time: 95.2s

Analysis

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Includes gradient fills for richer visual depth (great for the elephant body and skateboard texture)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Uses transforms (translate, rotate, scale) for positioning the elephant and tilting the skateboard
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders tusks – the model captured fine anatomical detail
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Renders trucks (the axle assemblies) – the model captured fine skateboard mechanical detail
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface
• Sky/atmosphere is present – the model created an outdoor scene

With 8,153 characters of SVG markup, this is a detailed output that balances richness with readability.

View raw SVG source

```xml ```

3. gemma4:31b-cloud

SVG size: 1240 characters
Complexity: Compact
Shape elements: 13
Distinct colors: 6
Raw response: 1898 characters
Generation time: 5.8s

Analysis

Notable SVG techniques used:

• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface

With 1,240 characters of SVG markup, this is a compact output – minimal but potentially efficient for web embedding.

View raw SVG source

```xml ```

4. gemma4:cloud

SVG size: 1155 characters
Complexity: Compact
Shape elements: 12
Distinct colors: 6
Raw response: 1934 characters
Generation time: 5.6s

Analysis

Notable SVG techniques used:

• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface

With 1,155 characters of SVG markup, this is a compact output – minimal but potentially efficient for web embedding.

View raw SVG source

```xml ```

5. glm-5.1:cloud

SVG size: 11862 characters
Complexity: Very high
Shape elements: 78
Distinct colors: 26
Raw response: 16000 characters
Generation time: 68.2s

Analysis

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Includes gradient fills for richer visual depth (great for the elephant body and skateboard texture)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Includes animation elements – the model attempted motion (e.g., rolling wheels, balancing)
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders tusks – the model captured fine anatomical detail
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Renders trucks (the axle assemblies) – the model captured fine skateboard mechanical detail
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface

With 11,862 characters of SVG markup, this is a detailed output that balances richness with readability.

View raw SVG source

```xml ```

6. glm-5.2:cloud

SVG size: 5364 characters
Complexity: Very high
Shape elements: 60
Distinct colors: 18
Raw response: 5394 characters
Generation time: 53.3s

Analysis

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Includes gradient fills for richer visual depth (great for the elephant body and skateboard texture)
• Uses transforms (translate, rotate, scale) for positioning the elephant and tilting the skateboard
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders tusks – the model captured fine anatomical detail
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Renders trucks (the axle assemblies) – the model captured fine skateboard mechanical detail
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface
• Sky/atmosphere is present – the model created an outdoor scene

With 5,364 characters of SVG markup, this is a detailed output that balances richness with readability.

View raw SVG source

```xml ```

7. gpt-oss:120b-cloud

SVG size: 1659 characters
Complexity: Medium
Shape elements: 20
Distinct colors: 7
Raw response: 2359 characters
Generation time: 9.1s

Analysis

Notable SVG techniques used:

• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface

With 1,659 characters of SVG markup, this is a compact output – minimal but potentially efficient for web embedding.

View raw SVG source

```xml ```

8. kimi-k2.6:cloud

SVG size: 7511 characters
Complexity: High
Shape elements: 30
Distinct colors: 15
Raw response: 22175 characters
Generation time: 49.9s

Analysis

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Includes gradient fills for richer visual depth (great for the elephant body and skateboard texture)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Uses transforms (translate, rotate, scale) for positioning the elephant and tilting the skateboard
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface
• Sky/atmosphere is present – the model created an outdoor scene

With 7,511 characters of SVG markup, this is a detailed output that balances richness with readability.

View raw SVG source

```xml ```

9. minimax-m2.7:cloud

SVG size: 3010 characters
Complexity: Medium
Shape elements: 24
Distinct colors: 8
Raw response: 3282 characters
Generation time: 37.1s

Analysis

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Includes gradient fills for richer visual depth (great for the elephant body and skateboard texture)
• Uses transforms (translate, rotate, scale) for positioning the elephant and tilting the skateboard
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders tusks – the model captured fine anatomical detail
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Renders trucks (the axle assemblies) – the model captured fine skateboard mechanical detail
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding

With 3,010 characters of SVG markup, this is a balanced output – enough detail to be recognizable without being bloated.

View raw SVG source

```xml ```

10. minimax-m3:cloud

SVG size: 5627 characters
Complexity: Very high
Shape elements: 63
Distinct colors: 23
Raw response: 6239 characters
Generation time: 99.6s

Analysis

Notable SVG techniques used:

• Renders actual text (labels, title, or decorative text)
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders tusks – the model captured fine anatomical detail
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Renders trucks (the axle assemblies) – the model captured fine skateboard mechanical detail
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface
• Sky/atmosphere is present – the model created an outdoor scene

With 5,627 characters of SVG markup, this is a detailed output that balances richness with readability.

View raw SVG source

```xml ```

11. nemotron-3-super:cloud

SVG size: 1995 characters
Complexity: Medium
Shape elements: 22
Distinct colors: 7
Raw response: 3705 characters
Generation time: 24.9s

Analysis

Notable SVG techniques used:

• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding

With 1,995 characters of SVG markup, this is a compact output – minimal but potentially efficient for web embedding.

View raw SVG source

```xml ```

12. nemotron-3-ultra:cloud

SVG size: 7239 characters
Complexity: High
Shape elements: 39
Distinct colors: 14
Raw response: 8408 characters
Generation time: 34.6s

Analysis

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Includes gradient fills for richer visual depth (great for the elephant body and skateboard texture)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Uses transforms (translate, rotate, scale) for positioning the elephant and tilting the skateboard
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders tusks – the model captured fine anatomical detail
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Renders trucks (the axle assemblies) – the model captured fine skateboard mechanical detail
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface

With 7,239 characters of SVG markup, this is a detailed output that balances richness with readability.

View raw SVG source

```xml ```

13. qwen3.5:397b-cloud

SVG size: 2604 characters
Complexity: Medium
Shape elements: 22
Distinct colors: 15
Raw response: 2651 characters
Generation time: 25.0s

Analysis

Notable SVG techniques used:

• Uses transforms (translate, rotate, scale) for positioning the elephant and tilting the skateboard
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Renders “elephant” in text/comments – the model captured the subject
• Renders a trunk – the model captured the elephant’s most distinctive feature
• Renders tusks – the model captured fine anatomical detail
• Renders ears – the model captured the elephant’s large ears (key for African vs Asian distinction)
• Renders legs – the model captured the elephant’s anatomy
• Renders a tail – the model captured full anatomical detail
• Renders eyes – the model gave the elephant character
• Renders “skateboard” in text/comments – the model captured the object
• Renders a deck (the board itself) – the model captured the skateboard’s main platform
• Renders wheels – the model captured the skateboard’s rolling mechanism (typically 4 wheels)
• Renders trucks (the axle assemblies) – the model captured fine skateboard mechanical detail
• Elephant-skateboard interaction is present – the model combined both concepts in one scene
• Motion/speed lines are present – the model captured the dynamic feel of skateboarding
• Ground/surface is present – the model placed the skateboard on a surface

With 2,604 characters of SVG markup, this is a balanced output – enough detail to be recognizable without being bloated.

View raw SVG source

```xml ```

Conclusion

That’s all 13 elephant-on-a-skateboard SVGs from the active Ollama cloud models. This whimsical-animal prompt tested a different skill set than our previous prompts: organic anatomy (elephant trunk, ears, legs), mechanical precision (skateboard deck, wheels, trucks), scale mismatch handling, and creative humor.

No single model is best at everything. A model that produces a stunning World Cup poster may produce a basic elephant, and vice versa. The best model for you depends on your specific use case:

• For maximum detail: look at the “Very high” complexity models
• For speed: look at the generation times in each section
• For anatomical accuracy: check which models captured the elephant’s trunk, ears, and tusks
• For skateboard accuracy: check which models rendered 4 wheels, a deck, and trucks
• For humor and whimsy: look for models that added motion lines, poses, or playful details
• For code quality: look at the raw SVG source – clean, well-structured code is easier to customize

Try the same prompt yourself with Ollama Cloud and see if you agree with our analysis. And check out the other posts in our SVG benchmark series:

• Duck on a bicycle
• Duck with a parachute
• Duck driving a jeep
• Cherry blossom trees
• Duck programmer debugging at 3am
• Baby shark fish
• Octopus playing chess
• FIFA World Cup 2026

Enjoyed this post? Never miss out on future posts by following us