Classic Mammoth Tank

One prompt. Different AI models.

A heavy twin-cannon Mammoth Tank inspired by classic Command & Conquer. Compare tracks, armor, missile pods and detail across independent Blender builds.

Download · Free

3D previews and build videos are free.

Drag to rotate · Scroll to zoom · Pinch on mobile

Compare the model runs

Same prompt. Different approaches. Inspect the results and the work behind them.

Comparison columns
Metrics
Editor’s pick
Iterations———1
Generated
Authoring time265 s253 s342.678 s458.034 s
Staged build time40.884 s27.783 s51.347 s34.89 s
Total tokens572,903674,641252,053295,505
Est. API cost$0.26279$0.0152893$0.90777$0.193274
Input560,736661,654241,108281,990
Cached input (included)544,640637,440227,840266,240
Output12,16712,98710,94513,515
Reasoning (included in output)3,7433,4481,8871,595
Reasoning efforthighhighhighhigh
Model / versiongpt-6-solgpt-6-lunagpt-6-astragpt-6.1-sol
SetupIntel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; CodexIntel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; CodexIntel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; CodexIntel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; Codex
Pricing checked2026-09-232026-09-232026-09-232026-10-01
Python execution0.044 s0.148 s29.396 s0.109 s
Build stages34233029
Objects386272895994
Vertices5,2373,0249,64013,843
Video duration54.267 s72.8 s64.933 s48.667 s
Blender file size261,154 bytes212,690 bytes527,827 bytes458,519 bytes
Downloads
Download · Free
Download · Free
Download · Free

Swipe between 4 models

Iterations
—
Generated
Authoring time
265 s
Staged build time
40.884 s
Total tokens
572,903
Est. API cost
$0.26279
Input
560,736
Cached input (included)
544,640
Output
12,167
Reasoning (included in output)
3,743
Reasoning effort
high
Model / version
gpt-6-sol
Setup
Intel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; Codex
Pricing checked
2026-09-23
Python execution
0.044 s
Build stages
34
Objects
386
Vertices
5,237
Video duration
54.267 s
Blender file size
261,154 bytes

Downloads

Download · Free
Iterations
—
Generated
Authoring time
253 s
Staged build time
27.783 s
Total tokens
674,641
Est. API cost
$0.0152893
Input
661,654
Cached input (included)
637,440
Output
12,987
Reasoning (included in output)
3,448
Reasoning effort
high
Model / version
gpt-6-luna
Setup
Intel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; Codex
Pricing checked
2026-09-23
Python execution
0.148 s
Build stages
23
Objects
272
Vertices
3,024
Video duration
72.8 s
Blender file size
212,690 bytes

Downloads

Download · Free
Iterations
—
Generated
Authoring time
342.678 s
Staged build time
51.347 s
Total tokens
252,053
Est. API cost
$0.90777
Input
241,108
Cached input (included)
227,840
Output
10,945
Reasoning (included in output)
1,887
Reasoning effort
high
Model / version
gpt-6-astra
Setup
Intel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; Codex
Pricing checked
2026-09-23
Python execution
29.396 s
Build stages
30
Objects
895
Vertices
9,640
Video duration
64.933 s
Blender file size
527,827 bytes

Downloads

Download · Free
Editor’s pick
Iterations
1
Generated
Authoring time
458.034 s
Staged build time
34.89 s
Total tokens
295,505
Est. API cost
$0.193274
Input
281,990
Cached input (included)
266,240
Output
13,515
Reasoning (included in output)
1,595
Reasoning effort
high
Model / version
gpt-6.1-sol
Setup
Intel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; Codex
Pricing checked
2026-10-01
Python execution
0.109 s
Build stages
29
Objects
994
Vertices
13,843
Video duration
48.667 s
Blender file size
458,519 bytes

Downloads

Compare the model runs

Answer and reasoning; reasoning is included in output.

One independent run per model
Tokens are text units. This counts the model’s answer and internal reasoning. More output does not automatically mean a better 3D model.
Estimated cost of this model run in USD at Standard API rates, including cached-input discounts. Local computing and shared coordination are excluded.
Measured agent time spent writing the Blender scene generator. Shows how long authoring took; the visible Blender build is measured separately.
Time for the visible Blender build, including pauses between stages. Shows the staged construction duration; Python execution and full video duration are measured separately.
Show in chart:
AnswerReasoning
16k12k8k4k0
12,167
3,743
8,424
12,987
3,448
9,539
10,945
9,058
13,515
11,920
GPT-6 Sol
GPT-6 Luna
GPT-6 Astra
GPT-6.1 SolEditor's pick
Context of the plotted runs
ModelsGPT-6 SolGPT-6 LunaGPT-6 AstraGPT-6.1 Sol
Model / versiongpt-6-solgpt-6-lunagpt-6-astragpt-6.1-sol
Reasoning efforthighhighhighhigh
IterationsNot recordedNot recordedNot recorded1
Pricing checked2026-09-232026-09-232026-09-232026-10-01
SetupIntel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; CodexIntel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; CodexIntel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; CodexIntel Core i9-12900KF · NVIDIA GeForce RTX 4060 Ti (16 GB) · 64 GB RAM · Windows 11 Pro · Blender 5.1.2; Codex

Original prompt

# Identical model brief

Create a detailed, immediately recognizable classic Mammoth Tank inspired by the original Command & Conquer in Blender 5.1.2. Exterior display model only: very heavy low armored hull, two broad caterpillar track assemblies with visible tread links and wheels, a substantial rotating-looking turret with twin parallel cannon barrels, twin missile pods, hatches, vents, headlights and restrained faction-color accents. Capture the chunky retro real-time-strategy silhouette and weathered military metal, in a polished stylized low-poly diorama on a small terrain display base. Cannons and tank front face negative Y. Keep the whole tank visible and centered. This is decorative fictional game art; no functional weapon design or technical internals. No downloaded assets. Decide the geometry, proportions, details and artistic interpretation independently.

Deliver scene.py in your assigned model directory. It must expose def build():, a Python generator using bpy that creates the scene from an empty scene and yields a short English stage description after each meaningful increment. Aim for 18–35 increments so an actual Blender window recording shows construction. No sleep, no rendering, no saving, no quitting Blender; the shared harness handles those. Module import must not create geometry. Use only bpy/mathutils/Python standard library. All scene geometry should fit approximately inside x/y [-9,9], z [0,11]. The subject faces negative Y. Assign material diffuse_color as well as node colors for visible material colors in solid viewport. Include a camera looking from (14,-19,13) toward (0,0,3) and suitable lights. Keep total objects below 2500 and avoid expensive modifiers. Use Blender 5.1 API. Generator execution happens in Blender UI via timer, so do not rely on area context for geometry. Do not modify UI or viewport; harness does that.

Own only your assigned model directory. Other workers may be present; do not revert or inspect their work or any other model's result. Do not use other AI models. Do not launch Blender, take over windows, install anything, or run an independent creative build. You may check Python syntax. Parent harness runs your actual Blender build and will report execution errors for repair. Write a concise design.txt identifying your intended result. End after these files are ready.

Same reasoning setting for all models: high. Each is a separate fresh agent. Timing separates model-authoring turnaround, actual Blender generator execution, visible staged build wall time, and common setup/QA. A single run is an exploratory comparison, not a statistically reliable benchmark.

Unlock the winning model