Ollama Cloud Models SVG Comparison: A Duck Driving a Bicycle
What happens when you ask 9 different state-of-the-art LLMs to draw the exact same thing? In this post, we sent the prompt Make an svg image of a duck driving a bicycle to every cloud model available in the local Ollama registry and compared the results side by side.
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.
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 of Results
| # | Model | SVG Size (chars) | Status |
|---|---|---|---|
| 1 | deepseek-v4-pro_cloud | 8044 | Generated |
| 2 | gemma4_31b-cloud | 1445 | Generated |
| 3 | glm-5.1_cloud | 14174 | Generated |
| 4 | glm-5.2_cloud | 5482 | Generated |
| 5 | kimi-k2.6_cloud | 6538 | Generated |
| 6 | minimax-m2.7_cloud | 3808 | Generated |
| 7 | nemotron-3-super_cloud | 1950 | Generated |
| 8 | nemotron-3-ultra_cloud | 11633 | Generated |
| 9 | qwen3.5_397b-cloud | 3492 | Generated |
| 10 | bjoernb/claude-opus-4-5:latest | - | Retired (410) |
| 11 | deepseek-v3.1:671b-cloud | - | Retired (410) |
| 12 | glm-5:cloud | - | Retired (410) |
| 13 | qwen3-vl:235b-cloud | - | Retired (410) |
9 out of 13 models successfully produced a valid SVG. The 4 retired models returned HTTP 410 Gone (they were removed from Ollama Cloud on 2026-07-15).
1. deepseek-v4-pro_cloud
SVG size: 8044 characters
Raw response: 8447 characters
Analysis
This SVG contains approximately 73 shape elements and uses 10 distinct colors.
Notable SVG techniques used:
- Uses
<defs>blocks for reusable components (a sign of well-structured SVG) - Uses
<use>references to reuse defined elements, reducing duplication - Uses transforms (translate, rotate, scale) for precise positioning
- Contains XML comments, showing the model’s internal documentation of its work
This is a moderately detailed SVG (5,000-10,000 characters), balancing visual richness with code clarity. The model included enough detail to make the scene recognizable while keeping the markup manageable. View raw SVG source
```xml ```
2. gemma4_31b-cloud
SVG size: 1445 characters
Raw response: 2015 characters
Analysis
This SVG contains approximately 16 shape elements and uses 7 distinct colors.
Notable SVG techniques used:
- Contains XML comments, showing the model’s internal documentation of its work
This is a compact SVG (under 5,000 characters) that uses minimal markup to convey the scene. The model prioritized efficiency and simplicity over elaborate detail. View raw SVG source
```xml ```
3. glm-5.1_cloud
SVG size: 14174 characters
Raw response: 16458 characters
Analysis
This SVG contains approximately 122 shape elements and uses 45 distinct colors.
Notable SVG techniques used:
- Uses
<defs>blocks for reusable components (a sign of well-structured SVG) - Includes gradient fills for richer visual depth
- Uses transforms (translate, rotate, scale) for precise positioning
- Contains XML comments, showing the model’s internal documentation of its work
With over 10,000 characters of SVG markup, this is one of the most detailed outputs in the comparison. The model invested significant effort in adding fine details, accessories, and scene elements. View raw SVG source
```xml ```
4. glm-5.2_cloud
SVG size: 5482 characters
Raw response: 5520 characters
Analysis
This SVG contains approximately 58 shape elements and uses 21 distinct colors.
Notable SVG techniques used:
- Uses transforms (translate, rotate, scale) for precise positioning
- Contains XML comments, showing the model’s internal documentation of its work
This is a moderately detailed SVG (5,000-10,000 characters), balancing visual richness with code clarity. The model included enough detail to make the scene recognizable while keeping the markup manageable. View raw SVG source
```xml ```
5. kimi-k2.6_cloud
SVG size: 6538 characters
Raw response: 23317 characters
Analysis
This SVG contains approximately 31 shape elements and uses 14 distinct colors.
Notable SVG techniques used:
- Uses
<defs>blocks for reusable components (a sign of well-structured SVG) - Includes gradient fills for richer visual depth
- Applies SVG filters (e.g., drop shadows) for visual effects
- Uses transforms (translate, rotate, scale) for precise positioning
- Contains XML comments, showing the model’s internal documentation of its work
This is a moderately detailed SVG (5,000-10,000 characters), balancing visual richness with code clarity. The model included enough detail to make the scene recognizable while keeping the markup manageable. View raw SVG source
```xml ```
6. minimax-m2.7_cloud
SVG size: 3808 characters
Raw response: 3999 characters
Analysis
This SVG contains approximately 30 shape elements and uses 11 distinct colors.
Notable SVG techniques used:
- Uses
<defs>blocks for reusable components (a sign of well-structured SVG) - Includes gradient fills for richer visual depth
- Applies SVG filters (e.g., drop shadows) for visual effects
- Uses transforms (translate, rotate, scale) for precise positioning
- Contains XML comments, showing the model’s internal documentation of its work
This is a compact SVG (under 5,000 characters) that uses minimal markup to convey the scene. The model prioritized efficiency and simplicity over elaborate detail. View raw SVG source
```xml ```
7. nemotron-3-super_cloud
SVG size: 1950 characters
Raw response: 1975 characters
Analysis
This SVG contains approximately 21 shape elements and uses 8 distinct colors.
Notable SVG techniques used:
- Contains XML comments, showing the model’s internal documentation of its work
This is a compact SVG (under 5,000 characters) that uses minimal markup to convey the scene. The model prioritized efficiency and simplicity over elaborate detail. View raw SVG source
```xml ```
8. nemotron-3-ultra_cloud
SVG size: 11633 characters
Raw response: 12821 characters
Analysis
This SVG contains approximately 92 shape elements and uses 26 distinct colors.
Notable SVG techniques used:
- Uses
<defs>blocks for reusable components (a sign of well-structured SVG) - Includes gradient fills for richer visual depth
- Uses transforms (translate, rotate, scale) for precise positioning
- Contains XML comments, showing the model’s internal documentation of its work
With over 10,000 characters of SVG markup, this is one of the most detailed outputs in the comparison. The model invested significant effort in adding fine details, accessories, and scene elements. View raw SVG source
```xml ```
9. qwen3.5_397b-cloud
SVG size: 3492 characters
Raw response: 3710 characters
Analysis
This SVG contains approximately 29 shape elements and uses 14 distinct colors.
Notable SVG techniques used:
- Uses transforms (translate, rotate, scale) for precise positioning
- Contains XML comments, showing the model’s internal documentation of its work
This is a compact SVG (under 5,000 characters) that uses minimal markup to convey the scene. The model prioritized efficiency and simplicity over elaborate detail. View raw SVG source
```xml ```
Retired Models
The following models were listed in the Ollama registry but returned HTTP 410 Gone when queried. They were retired from Ollama Cloud and are no longer available for inference:
bjoernb/claude-opus-4-5:latest– glm-4.7 was retired at 2026-07-15deepseek-v3.1:671b-cloud– deepseek-v3.1:671b was retired at 2026-07-15glm-5:cloud– glm-5 was retired at 2026-07-15qwen3-vl:235b-cloud– qwen3-vl:235b was retired at 2026-06-16
Conclusion
This comparison shows that 9 out of 13 active Ollama cloud models can generate valid SVG artwork from a natural language prompt. The results vary dramatically in complexity, style, and technique:
- Simplest output: gemma4:31b-cloud produced a compact 1,445-character SVG with basic shapes
- Most detailed output: glm-5.1:cloud generated a 14,000+ character SVG with rich scene elements
- Most technically advanced: deepseek-v4-pro:cloud used
<defs>,<use>, transforms, and motion lines - Fastest response: gemma4:31b-cloud completed in just 7 seconds
All SVGs were generated with zero post-processing – what you see is exactly what each model produced. The variety of approaches (from minimalist geometric to richly detailed scenes) highlights how different training data and model architectures lead to distinctly different creative outputs.
Links
Enjoyed this post? Never miss out on future posts by following us
