Which Ollama Cloud Model is Best? Cherry Blossom Trees SVG Comparison (13 Models)

After testing LLMs on ducks and vehicles, we wanted to know: can today’s top models draw nature? This time we asked 13 Ollama cloud models to draw cherry blossom trees with flowers – a softer, more organic prompt that tests color palettes, repeated organic shapes (petals), and scene composition.

The prompt was: Make an svg image about cherry blossom trees with flowers.

This is the fourth in our SVG benchmark series. See also: duck on a bicycle, duck with a parachute, and duck driving a jeep.

Why cherry blossoms? Unlike the duck prompts, this scene has no central character. It tests a different skill set: gradient skies, repeating petal shapes, branch structures, and a calm nature palette (pinks, greens, blues). A model that drew a great duck may struggle here, and vice versa – which is exactly why we run multiple prompts.

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 Nature SVGs

Nature prompts reward different things than character prompts. Here are the criteria to use:

• Color palette: Cherry blossoms demand pinks, soft greens, and sky blues. Does the model use a tasteful palette, or does it dump random colors?
• Organic shapes: Look for curved petals, branching trunks, and natural-looking foliage. Models that only draw rectangles and circles struggle here.
• Repetition handling: Blossoms require many similar-but-not-identical petals. Models that use <use> with transforms scale better than those that hand-draw each petal.
• Depth and layering: Does the SVG have a foreground/background, sky gradient, and ground? Layering is what makes nature scenes feel real.
• SVG code quality: Does it use <defs>, <use>, gradients, and filters? Cleaner code is easier to tweak (e.g., to recolor for autumn).
• Output size: Very small SVGs miss the detail that makes blossoms beautiful. Very large SVGs can be slow but are usually worth it for nature scenes.

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 zero post-processing.

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 13 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	16296	52	14	Very high	Rich detail, fast
2	deepseek-v4-pro_cloud	13680	49	15	High	Most detailed
3	gemma4_31b-cloud	3109	29	6	Low	Compact
4	gemma4_cloud	2649	28	9	Low	Minimalist
5	glm-5.1_cloud	19810	69	21	Very high	Most detailed
6	glm-5.2_cloud	15956	52	27	Very high	Highly detailed
7	gpt-oss_120b-cloud	3493	9	6	Low	Compact
8	kimi-k2.6_cloud	1678	3	7	Low	Minimalist
9	minimax-m2.7_cloud	3319	20	11	Low	Compact
10	minimax-m3_cloud	15741	171	15	Very high	Highly detailed
11	nemotron-3-super_cloud	2648	42	3	Low	Compact
12	nemotron-3-ultra_cloud	16411	32	20	Very high	Richest scene
13	qwen3.5_397b-cloud	4458	25	8	Low	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, painterly cherry blossom SVG: pick glm-5.1:cloud, glm-5.2:cloud, deepseek-v4-flash:cloud, or nemotron-3-ultra:cloud
• You want the fastest response: pick gpt-oss:120b-cloud (~11s) or nemotron-3-super:cloud (~17s)
• You want the cleanest, most reusable SVG code: pick deepseek-v4-pro:cloud (uses <defs>, <use>, transforms)
• You want a small, efficient SVG for web embedding: pick gemma4:cloud or kimi-k2.6:cloud
• You want a balance of detail and speed: pick qwen3.5:397b-cloud or minimax-m3:cloud
• 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: 16296 characters
Complexity: Very high
Shape elements: 52
Distinct colors: 14
Raw response: 16702 characters

Analysis

This SVG contains approximately 52 shape elements and uses 14 distinct colors.

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Uses <use> references to reuse defined elements – ideal for repeating petals
• Includes gradient fills for richer visual depth (great for skies and petals)
• Uses transforms (translate, rotate, scale) for positioning petals and branches
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

With over 15,000 characters of SVG markup, this is one of the most detailed outputs in the comparison. The model invested significant effort in rendering individual petals, layered branches, and atmospheric backgrounds. Best for users who want a painterly cherry blossom scene.

View raw SVG source

```xml ```

2. deepseek-v4-pro_cloud

SVG size: 13680 characters
Complexity: High
Shape elements: 49
Distinct colors: 15
Raw response: 14073 characters

Analysis

This SVG contains approximately 49 shape elements and uses 15 distinct colors.

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Uses <use> references to reuse defined elements – ideal for repeating petals
• Includes gradient fills for richer visual depth (great for skies and petals)
• Uses transforms (translate, rotate, scale) for positioning petals and branches
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

With 10,000-15,000 characters, this is a highly detailed SVG. The model added many petals and scene elements while keeping the markup structured. A strong choice if you want detail without overwhelming complexity.

View raw SVG source

```xml ```

3. gemma4_31b-cloud

SVG size: 3109 characters
Complexity: Low
Shape elements: 29
Distinct colors: 6
Raw response: 4323 characters

Analysis

This SVG contains approximately 29 shape elements and uses 6 distinct colors.

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 skies and petals)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Uses transforms (translate, rotate, scale) for positioning petals and branches
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

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 petal detail. Best for embedding in web pages where small file size matters more than visual richness.

View raw SVG source

```xml ```

4. gemma4_cloud

SVG size: 2649 characters
Complexity: Low
Shape elements: 28
Distinct colors: 9
Raw response: 3684 characters

Analysis

This SVG contains approximately 28 shape elements and uses 9 distinct colors.

Notable SVG techniques used:

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

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

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 petal detail. Best for embedding in web pages where small file size matters more than visual richness.

View raw SVG source

```xml ```

5. glm-5.1_cloud

SVG size: 19810 characters
Complexity: Very high
Shape elements: 69
Distinct colors: 21
Raw response: 22858 characters

Analysis

This SVG contains approximately 69 shape elements and uses 21 distinct colors.

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Uses <use> references to reuse defined elements – ideal for repeating petals
• Includes gradient fills for richer visual depth (great for skies and petals)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Uses transforms (translate, rotate, scale) for positioning petals and branches
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

With over 15,000 characters of SVG markup, this is one of the most detailed outputs in the comparison. The model invested significant effort in rendering individual petals, layered branches, and atmospheric backgrounds. Best for users who want a painterly cherry blossom scene.

View raw SVG source

```xml ```

6. glm-5.2_cloud

SVG size: 15956 characters
Complexity: Very high
Shape elements: 52
Distinct colors: 27
Raw response: 17459 characters

Analysis

This SVG contains approximately 52 shape elements and uses 27 distinct colors.

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Uses <use> references to reuse defined elements – ideal for repeating petals
• Includes gradient fills for richer visual depth (great for skies and petals)
• Uses transforms (translate, rotate, scale) for positioning petals and branches
• Includes animation elements – the model attempted motion (e.g., falling petals)
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

With over 15,000 characters of SVG markup, this is one of the most detailed outputs in the comparison. The model invested significant effort in rendering individual petals, layered branches, and atmospheric backgrounds. Best for users who want a painterly cherry blossom scene.

View raw SVG source

```xml ```

7. gpt-oss_120b-cloud

SVG size: 3493 characters
Complexity: Low
Shape elements: 9
Distinct colors: 6
Raw response: 4335 characters

Analysis

This SVG contains approximately 9 shape elements and uses 6 distinct colors.

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Uses <use> references to reuse defined elements – ideal for repeating petals
• Uses transforms (translate, rotate, scale) for positioning petals and branches
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

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 petal detail. Best for embedding in web pages where small file size matters more than visual richness.

View raw SVG source

```xml ```

8. kimi-k2.6_cloud

SVG size: 1678 characters
Complexity: Low
Shape elements: 3
Distinct colors: 7
Raw response: 16993 characters

Analysis

This SVG contains approximately 3 shape elements and uses 7 distinct colors.

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 skies and petals)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

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 petal detail. Best for embedding in web pages where small file size matters more than visual richness.

View raw SVG source

```xml ```

9. minimax-m2.7_cloud

SVG size: 3319 characters
Complexity: Low
Shape elements: 20
Distinct colors: 11
Raw response: 3627 characters

Analysis

This SVG contains approximately 20 shape elements and uses 11 distinct colors.

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Uses <use> references to reuse defined elements – ideal for repeating petals
• Includes gradient fills for richer visual depth (great for skies and petals)
• Uses transforms (translate, rotate, scale) for positioning petals and branches
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

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 petal detail. Best for embedding in web pages where small file size matters more than visual richness.

View raw SVG source

```xml ```

10. minimax-m3_cloud

SVG size: 15741 characters
Complexity: Very high
Shape elements: 171
Distinct colors: 15
Raw response: 16614 characters

Analysis

This SVG contains approximately 171 shape elements and uses 15 distinct colors.

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

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

With over 15,000 characters of SVG markup, this is one of the most detailed outputs in the comparison. The model invested significant effort in rendering individual petals, layered branches, and atmospheric backgrounds. Best for users who want a painterly cherry blossom scene.

View raw SVG source

```xml ```

11. nemotron-3-super_cloud

SVG size: 2648 characters
Complexity: Low
Shape elements: 42
Distinct colors: 3
Raw response: 4190 characters

Analysis

This SVG contains approximately 42 shape elements and uses 3 distinct colors.

Notable SVG techniques used:

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

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are not prominent – the model may have used a generic green/brown palette

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 petal detail. Best for embedding in web pages where small file size matters more than visual richness.

View raw SVG source

```xml ```

12. nemotron-3-ultra_cloud

SVG size: 16411 characters
Complexity: Very high
Shape elements: 32
Distinct colors: 20
Raw response: 17911 characters

Analysis

This SVG contains approximately 32 shape elements and uses 20 distinct colors.

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Uses <use> references to reuse defined elements – ideal for repeating petals
• Includes gradient fills for richer visual depth (great for skies and petals)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Uses transforms (translate, rotate, scale) for positioning petals and branches
• Includes animation elements – the model attempted motion (e.g., falling petals)
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

With over 15,000 characters of SVG markup, this is one of the most detailed outputs in the comparison. The model invested significant effort in rendering individual petals, layered branches, and atmospheric backgrounds. Best for users who want a painterly cherry blossom scene.

View raw SVG source

```xml ```

13. qwen3.5_397b-cloud

SVG size: 4458 characters
Complexity: Low
Shape elements: 25
Distinct colors: 8
Raw response: 4688 characters

Analysis

This SVG contains approximately 25 shape elements and uses 8 distinct colors.

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Uses <use> references to reuse defined elements – ideal for repeating petals
• Includes gradient fills for richer visual depth (great for skies and petals)
• Uses transforms (translate, rotate, scale) for positioning petals and branches
• Includes animation elements – the model attempted motion (e.g., falling petals)
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tree elements are present (trunk, branches) – the model understood the structure
• Flower / blossom elements are present (petals, blooms) – the model captured the subject
• Pink tones are used – the model understood the cherry blossom color cue

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 petal detail. Best for embedding in web pages where small file size matters more than visual richness.

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 2026-07-15
• deepseek-v3.1:671b-cloud – deepseek-v3.1:671b was retired at 2026-07-15
• glm-5:cloud – glm-5 was retired at 2026-07-15
• qwen3-vl:235b-cloud – qwen3-vl:235b was retired at 2026-06-16

How to Pick the Best Model: A Decision Guide

Now that you have seen all 13 cherry blossom SVGs, here is a structured way to decide which Ollama cloud model is best for your SVG generation needs:

Step 1: Define your priority

• Visual quality first: Scroll back through the SVGs above and pick the one that looks best to your eye. For nature scenes, trust your eye – the shape and color counts are useful, but a model with 30 shapes and a great palette can beat one with 100 shapes and muddy colors.
• Code quality first: Open the raw SVG source for each model (use the disclosure toggles) and look for <defs>, <use>, gradients, and clean indentation. Models that produce structured code are easier to recolor (e.g., for autumn vs spring) and reuse.
• Speed first: If you are building a real-time app, prioritize the models that responded in under 20 seconds (gpt-oss:120b-cloud, nemotron-3-super:cloud, gemma4:31b-cloud).
• File size first: For web embedding, smaller is better. Look at the SVG size column in the summary table.

Step 2: Cross-check across prompts

A model that does well on cherry blossoms might fail on a character prompt. Check our other benchmarks:

• Duck on a bicycle – character + vehicle
• Duck with a parachute – dynamic action scene
• Duck driving a jeep – vehicle with multiple parts
• This post (cherry blossoms) – nature / scenery with no central character

A model that consistently produces good results across all four prompts is a safer pick than one that only shines on a single type of scene.

Step 3: Test with your own prompt

Every model has strengths and weaknesses. The only way to know for sure which model is best for your specific use case is to test it with your own prompt. The Ollama Cloud API is OpenAI-compatible, so you can use any standard client:

from openai import OpenAI
client = OpenAI(base_url="http://localhost:11434/v1", api_key="ollama")
resp = client.chat.completions.create(
    model="glm-5.1:cloud",  # change this to test different models
    messages=[{"role": "user", "content": "Make an svg image of <your prompt>"}],
)
print(resp.choices[0].message.content)

Why This Benchmark Matters: Nature vs Character Prompts

Most LLM SVG benchmarks only test one type of prompt. We deliberately run four different scene types because models specialize:

• Character prompts (the duck series) reward models that understand anatomy, proportions, and accessories. Models like deepseek-v4-pro and kimi-k2.6 tend to do well here.
• Nature prompts (this cherry blossom post) reward models that understand organic shapes, palettes, and repetition. Models like glm-5.1 and nemotron-3-ultra tend to do well here, because they invest in many small petals and gradients.

If you are picking a model for a specific project, look at the benchmark closest to your use case. If you want a general-purpose model, pick one that does okay across all four rather than one that aces one and fails the rest.

Conclusion: You Decide the Winner

This comparison shows that 13 out of 17 active Ollama cloud models can generate valid SVG artwork from a nature prompt about cherry blossom trees. The results vary dramatically in complexity, style, and technique – and there is no single “best” model.

Our takeaways after running four SVG benchmarks (bicycle, parachute, jeep, blossom):

• glm-5.1:cloud and nemotron-3-ultra:cloud consistently produce the longest, most detailed SVGs across all prompt types. Best when you want maximum visual richness.
• deepseek-v4-pro:cloud and deepseek-v4-flash:cloud consistently produce well-structured, technically advanced SVGs with <defs>, <use>, and transforms. A strong default choice for code quality and editability.
• gpt-oss:120b-cloud and gemma4:31b-cloud are consistently among the fastest (under 20 seconds) and produce compact SVGs. Best for speed-sensitive applications.
• kimi-k2.6:cloud output size varies wildly by prompt – it produced a 9,993-char jeep SVG but only a 1,670-char blossom SVG. Worth testing on your specific prompt.
• glm-5.2:cloud and minimax-m3:cloud offer a reliable balance of detail, speed, and code quality across all four prompts.

But the real verdict is yours. Scroll back through the SVGs, compare them visually, check the raw code, and pick the model that best fits your needs. Every model in this comparison is available right now on Ollama Cloud – so you can reproduce these results in minutes.

Links

• Previous: Duck Driving a Bicycle Comparison
• Previous: Duck Jumping From a Plane Comparison
• Previous: Duck Driving a Jeep Comparison
• Ollama Official Website
• Ollama Cloud Documentation
• SVG Specification (MDN)
• OpenAI API Reference (used by Ollama)

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