Which Ollama Cloud Model is Best? Octopus Playing Chess SVG Comparison (13 Models)

After testing LLMs on ducks, vehicles, dev scenes, marine life, and nature, we wanted to know: can today’s top models draw a complex, imaginative scene? This time we asked 13 Ollama cloud models to draw an octopus playing chess – a prompt that tests organic anatomy, geometric precision, scene composition, and creative concept combination.

The prompt was: Make svg image of an octopus playing chess

This is the seventh 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, and baby shark fish.

Why an octopus playing chess? This prompt is the ultimate stress test for SVG generation because it combines multiple hard problems: (1) organic anatomy – an octopus has 8 tentacles with suction cups, a bulbous head/mantle, and eyes, all requiring complex <path> curves, (2) geometric precision – a chess board needs exactly 64 squares in an 8x8 grid with alternating colors, plus 32 pieces with specific shapes (pawn, rook, knight, bishop, queen, king), (3) scene composition – the octopus must be positioned AT the chess board, with tentacles reaching toward pieces, (4) concept combination – the model must understand both “octopus” and “chess” and combine them logically, (5) scale and proportion – the octopus should be larger than the chess pieces but not dominate the entire scene. A model that draws a great shark may fail here because chess requires precise geometry, not just organic curves.

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 Complex SVG Scenes

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

• Octopus anatomy: Does the SVG have a head/mantle, 8 tentacles, and eyes? Or is it a generic blob? Tentacles are the hardest part – they require complex bezier curves.
• Chess accuracy: Does the SVG have an 8x8 board with 64 squares? Are there recognizable chess pieces (pawn, rook, knight, bishop, queen, king)? Chess pieces have specific silhouettes.
• Interaction: Is the octopus actually interacting with the chess board? Are tentacles reaching toward pieces? This tests whether the model understood “playing” not just “octopus” and “chess” separately.
• Scene composition: Is the octopus positioned at the board? Is the scale reasonable? A good scene has the octopus on one side and the board in front of it.
• Creative interpretation: Did the model add underwater elements (bubbles, water) to contextualize the octopus? Or did it treat it as a dry-land chess game?
• SVG code quality: Does it use <defs>, <use>, gradients, and clean structure? Better code is easier to tweak (e.g., to recolor or animate).

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.

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 octopus chess scene 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 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: 14766 characters
Complexity: Very high
Shape elements: 99
Distinct colors: 17
Raw response: 15233 characters
Generation time: 233.8s

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 octopus body and chess pieces)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Uses <use> references to reuse defined elements – ideal for repeating chess squares or tentacles
• Uses transforms (translate, rotate, scale) for positioning the octopus and chess pieces
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Eyes are present – the model gave the octopus character
• Suction cups are present – the model captured fine anatomical detail
• Chess board is present – the model captured the game setting
• Pawn pieces are present – the model rendered chess pieces
• Rook pieces are present – the model rendered chess pieces
• Knight pieces are present – the model rendered chess pieces
• Queen piece is present – the model rendered the most powerful chess piece
• King piece is present – the model rendered the most important chess piece
• Tentacle-chess interaction is present – the model captured the octopus playing chess
• Bubbles are present – the model created an underwater atmosphere
• Underwater scene is present – the model understood the octopus’s habitat

With 14,766 characters of SVG markup, this is a detailed output that balances richness with readability.

2. deepseek-v4-pro:cloud

SVG size: 13058 characters
Complexity: Very high
Shape elements: 81
Distinct colors: 13
Raw response: 13533 characters
Generation time: 148.1s

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 octopus body and chess pieces)
• Uses <use> references to reuse defined elements – ideal for repeating chess squares or tentacles
• Uses transforms (translate, rotate, scale) for positioning the octopus and chess pieces
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Eyes are present – the model gave the octopus character
• Suction cups are present – the model captured fine anatomical detail
• Chess board is present – the model captured the game setting
• Pawn pieces are present – the model rendered chess pieces
• Rook pieces are present – the model rendered chess pieces
• Knight pieces are present – the model rendered chess pieces
• Bishop pieces are present – the model rendered chess pieces
• Queen piece is present – the model rendered the most powerful chess piece
• King piece is present – the model rendered the most important chess piece
• Tentacle-chess interaction is present – the model captured the octopus playing chess

With 13,058 characters of SVG markup, this is a detailed output that balances richness with readability.

3. gemma4:31b-cloud

SVG size: 2394 characters
Complexity: Medium
Shape elements: 27
Distinct colors: 13
Raw response: 3367 characters
Generation time: 27.1s

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:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Eyes are present – the model gave the octopus character
• Suction cups are present – the model captured fine anatomical detail
• Chess board is present – the model captured the game setting
• Tentacle-chess interaction is present – the model captured the octopus playing chess
• Bubbles are present – the model created an underwater atmosphere

With 2,394 characters of SVG markup, this is a balanced output – enough detail to be recognizable without being bloated.

4. gemma4:cloud

SVG size: 2823 characters
Complexity: High
Shape elements: 36
Distinct colors: 9
Raw response: 3414 characters
Generation time: 16.1s

Analysis

Notable SVG techniques used:

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

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Eyes are present – the model gave the octopus character
• Chess board is present – the model captured the game setting
• Tentacle-chess interaction is present – the model captured the octopus playing chess

With 2,823 characters of SVG markup, this is a balanced output – enough detail to be recognizable without being bloated.

5. glm-5.1:cloud

SVG size: 28577 characters
Complexity: Very high
Shape elements: 224
Distinct colors: 22
Raw response: 35887 characters
Generation time: 107.2s

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 octopus body and chess pieces)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Uses transforms (translate, rotate, scale) for positioning the octopus and chess pieces
• Renders actual text (labels, title, or decorative text)
• Includes animation elements – the model attempted motion (e.g., tentacle movement, floating)
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Eyes are present – the model gave the octopus character
• Suction cups are present – the model captured fine anatomical detail
• Chess board is present – the model captured the game setting
• Pawn pieces are present – the model rendered chess pieces
• Rook pieces are present – the model rendered chess pieces
• Knight pieces are present – the model rendered chess pieces
• Bishop pieces are present – the model rendered chess pieces
• Queen piece is present – the model rendered the most powerful chess piece
• King piece is present – the model rendered the most important chess piece
• Chess board squares are colored – the model rendered the checkered pattern
• Tentacle-chess interaction is present – the model captured the octopus playing chess
• Bubbles are present – the model created an underwater atmosphere
• Underwater scene is present – the model understood the octopus’s habitat

With over 28,577 characters of SVG markup, this is one of the most detailed outputs in the comparison. The model invested significant effort in rendering the octopus anatomy, chess board, pieces, and scene composition.

6. glm-5.2:cloud

SVG size: 10904 characters
Complexity: Very high
Shape elements: 89
Distinct colors: 14
Raw response: 10975 characters
Generation time: 80.3s

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

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Eyes are present – the model gave the octopus character
• Suction cups are present – the model captured fine anatomical detail
• Chess board is present – the model captured the game setting
• Pawn pieces are present – the model rendered chess pieces
• Rook pieces are present – the model rendered chess pieces
• Knight pieces are present – the model rendered chess pieces
• Queen piece is present – the model rendered the most powerful chess piece
• King piece is present – the model rendered the most important chess piece
• Chess board squares are colored – the model rendered the checkered pattern
• Tentacle-chess interaction is present – the model captured the octopus playing chess
• Bubbles are present – the model created an underwater atmosphere
• Underwater scene is present – the model understood the octopus’s habitat

With 10,904 characters of SVG markup, this is a detailed output that balances richness with readability.

7. gpt-oss:120b-cloud

SVG size: 5904 characters
Complexity: Medium
Shape elements: 24
Distinct colors: 7
Raw response: 6568 characters
Generation time: 21.1s

Analysis

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 chess squares or tentacles
• Uses transforms (translate, rotate, scale) for positioning the octopus and chess pieces
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Eyes are present – the model gave the octopus character
• Chess board is present – the model captured the game setting
• King piece is present – the model rendered the most important chess piece
• Chess board squares are colored – the model rendered the checkered pattern
• Tentacle-chess interaction is present – the model captured the octopus playing chess

With 5,904 characters of SVG markup, this is a detailed output that balances richness with readability.

8. kimi-k2.6:cloud

SVG size: 6600 characters
Complexity: Medium
Shape elements: 16
Distinct colors: 9
Raw response: 29480 characters
Generation time: 46.5s

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

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Eyes are present – the model gave the octopus character
• Chess board is present – the model captured the game setting
• Tentacle-chess interaction is present – the model captured the octopus playing chess
• Bubbles are present – the model created an underwater atmosphere
• Underwater scene is present – the model understood the octopus’s habitat

With 6,600 characters of SVG markup, this is a detailed output that balances richness with readability.

9. minimax-m2.7:cloud

SVG size: 6498 characters
Complexity: Very high
Shape elements: 66
Distinct colors: 9
Raw response: 6657 characters
Generation time: 78.7s

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

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Eyes are present – the model gave the octopus character
• Chess board is present – the model captured the game setting
• Pawn pieces are present – the model rendered chess pieces
• King piece is present – the model rendered the most important chess piece
• Chess board squares are colored – the model rendered the checkered pattern
• Tentacle-chess interaction is present – the model captured the octopus playing chess

With 6,498 characters of SVG markup, this is a detailed output that balances richness with readability.

10. minimax-m3:cloud

SVG size: 16078 characters
Complexity: Very high
Shape elements: 162
Distinct colors: 16
Raw response: 16183 characters
Generation time: 37.4s

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 octopus body and chess pieces)
• Uses transforms (translate, rotate, scale) for positioning the octopus and chess pieces
• Renders actual text (labels, title, or decorative text)
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Eyes are present – the model gave the octopus character
• Suction cups are present – the model captured fine anatomical detail
• Chess board is present – the model captured the game setting
• Pawn pieces are present – the model rendered chess pieces
• Rook pieces are present – the model rendered chess pieces
• Knight pieces are present – the model rendered chess pieces
• Bishop pieces are present – the model rendered chess pieces
• Queen piece is present – the model rendered the most powerful chess piece
• King piece is present – the model rendered the most important chess piece
• Chess board squares are colored – the model rendered the checkered pattern
• Tentacle-chess interaction is present – the model captured the octopus playing chess
• Bubbles are present – the model created an underwater atmosphere
• Underwater scene is present – the model understood the octopus’s habitat

With over 16,078 characters of SVG markup, this is one of the most detailed outputs in the comparison. The model invested significant effort in rendering the octopus anatomy, chess board, pieces, and scene composition.

11. nemotron-3-super:cloud

SVG size: 2155 characters
Complexity: Compact
Shape elements: 14
Distinct colors: 6
Raw response: 2730 characters
Generation time: 21.8s

Analysis

Notable SVG techniques used:

• Uses <defs> blocks for reusable components (a sign of well-structured, editable SVG)
• Renders actual text (labels, title, or decorative text)
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Chess board is present – the model captured the game setting
• Rook pieces are present – the model rendered chess pieces
• Bishop pieces are present – the model rendered chess pieces
• Queen piece is present – the model rendered the most powerful chess piece
• King piece is present – the model rendered the most important chess piece
• Chess board squares are colored – the model rendered the checkered pattern
• Tentacle-chess interaction is present – the model captured the octopus playing chess

With 2,155 characters of SVG markup, this is a balanced output – enough detail to be recognizable without being bloated.

12. nemotron-3-ultra:cloud

SVG size: 21145 characters
Complexity: Very high
Shape elements: 148
Distinct colors: 19
Raw response: 21413 characters
Generation time: 111.5s

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 octopus body and chess pieces)
• Applies SVG filters (e.g., drop shadows, blurs) for atmospheric effects
• Uses <use> references to reuse defined elements – ideal for repeating chess squares or tentacles
• Uses transforms (translate, rotate, scale) for positioning the octopus and chess pieces
• Renders actual text (labels, title, or decorative text)
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Eyes are present – the model gave the octopus character
• Suction cups are present – the model captured fine anatomical detail
• Chess board is present – the model captured the game setting
• Pawn pieces are present – the model rendered chess pieces
• Rook pieces are present – the model rendered chess pieces
• Knight pieces are present – the model rendered chess pieces
• Bishop pieces are present – the model rendered chess pieces
• Queen piece is present – the model rendered the most powerful chess piece
• King piece is present – the model rendered the most important chess piece
• Chess board squares are colored – the model rendered the checkered pattern
• Tentacle-chess interaction is present – the model captured the octopus playing chess
• Underwater scene is present – the model understood the octopus’s habitat

With over 21,145 characters of SVG markup, this is one of the most detailed outputs in the comparison. The model invested significant effort in rendering the octopus anatomy, chess board, pieces, and scene composition.

13. qwen3.5:397b-cloud

SVG size: 7464 characters
Complexity: Very high
Shape elements: 89
Distinct colors: 13
Raw response: 7826 characters
Generation time: 61.5s

Analysis

Notable SVG techniques used:

• Includes gradient fills for richer visual depth (great for the octopus body and chess pieces)
• Uses transforms (translate, rotate, scale) for positioning the octopus and chess pieces
• Contains XML comments, showing the model’s internal documentation of its work

Concept adherence:

• Tentacles are present – the model captured the octopus’s defining feature
• Octopus is explicitly referenced in element IDs/comments – the model understood the subject
• Octopus head/mantle is present – the model captured cephalopod anatomy
• Eyes are present – the model gave the octopus character
• Chess board is present – the model captured the game setting
• Pawn pieces are present – the model rendered chess pieces
• Rook pieces are present – the model rendered chess pieces
• King piece is present – the model rendered the most important chess piece
• Chess board squares are colored – the model rendered the checkered pattern
• Tentacle-chess interaction is present – the model captured the octopus playing chess

With 7,464 characters of SVG markup, this is a detailed output that balances richness with readability.

Conclusion

That’s all 13 octopus chess SVGs from the active Ollama cloud models. This creative-combination prompt tested a different skill set than our previous prompts: organic anatomy (tentacles), geometric precision (chess board), scene composition, and concept combination.

No single model is best at everything. A model that produces a stunning cherry blossom may produce a basic octopus, 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 code quality: look at the raw SVG source – clean, well-structured code is easier to customize
• For concept adherence: check which models captured the octopus’s tentacles, the chess board, and the interaction between them

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

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