Build a Pan Tricycle Chef with GPT-6 Astra and V2Fun

Overview

This case turns a whimsical character-and-vehicle reference into a complete 3D asset: a round-faced chef seated on a tricycle whose body resembles a black frying pan. The visual evidence shows three stages—modeling references, an exploded component study, and the final assembled model.

The workflow uses GPT-6 Astra to interpret the reference, plan the component hierarchy, and coordinate assembly. V2Fun supplies the detailed chef mesh, while regular mechanical parts are represented procedurally. This division follows the broader GPT-6 Astra and V2Fun model-collaboration workflow, in which organic forms and structured geometry are handled according to their different modeling needs.

Tools Used

• GPT-6 Astra: reference interpretation, component planning, spatial reasoning, and assembly calibration.
• V2Fun: generation of the textured chef character shown in the exploded study. Learn more about the platform's AI 3D model generator.
• Three.js / procedural construction: the structural frame, running gear, pan-shaped body and seat, mechanical equipment, controls, and supports identified in the process image.
• Visual references: a complete concept image, an isolated character reference, and an AI-inferred front, side, and rear study. The single-view and multi-view input guide explains why additional angles can reduce uncertainty around hidden forms.
• GLB: the supplied final delivery format; see the GLB and 3D export format guide for general format-selection context.

Step 1 — Interpret the Design Intent

GPT-6 Astra identifies two interdependent subjects: the chef and the frying-pan tricycle. The reference establishes the chef's serious expression, white uniform, apron, padded seat, handlebar grip, and foot placement. It also defines the vehicle's shallow pan body, rearward pan handle, red tubular frame, cream wheel hubs, three-wheel layout, and paired short exhaust pipes.

Step 2 — Prepare Reference Views

The modeling study retains the complete concept, isolates the chef, and adds an AI-inferred three-view study of the combined design. These inferred views provide working guidance for the front, side, and rear relationships, but they are interpretations rather than independently photographed evidence.

Step 3 — Decompose the Asset

The exploded study separates the model into six functional groups: character, structural frame, running gear, body and seat, mechanical equipment, and controls and supports. The chef is explicitly labeled as V2Fun-generated 3D, while the other groups are labeled as Three.js procedural components.

Step 4 — Generate the Character

V2Fun generates the isolated textured chef, preserving the visible facial volume, chef's hat, clothing layers, folds, seated pose, boots, and forward-reaching arms. Separating the character allows the organic form to be handled independently from the vehicle's regular mechanical construction.

Step 5 — Build the Mechanical Components

The vehicle is constructed from clearly bounded procedural forms. The exploded view shows a red frame, three wheels, a shallow black pan body with seat and rear handle, short cylindrical equipment, and a handlebar-and-support assembly. These elements correspond to the prompt's emphasis on cylinders, discs, rectangular plates, straight rods, and a limited number of curved tubes.

Step 6 — Calibrate Contact and Assembly

On the agent side, the generated character and procedural vehicle parts are brought together by adjusting scale, rotation, and position. The final render shows the chef seated over the pan body, both hands aligned with the handlebars, and both feet placed near the external supports. The frame, wheels, seat, rear handle, and small mechanical parts are combined into one visually coherent assembly.

Workflow

Concept reference → Isolate character → Infer supporting views → Decompose components → Generate character with V2Fun → Build regular parts procedurally → Calibrate contacts → Assemble → Export GLB

Prompt

Create a complete standalone asset featuring one round-faced, heavyset chef seriously driving a frying-pan-shaped tricycle. He wears a puffy chef's hat, white chef's uniform, and apron. The vehicle has a shallow black frying-pan body, a rearward pan handle forming a tail beam, an independent padded seat, one steering wheel at the front, two wheels at the rear, a fully visible tubular frame, and two short cylindrical exhaust pipes. The chef grips curved handlebars with both hands and rests his feet on external foot supports. Contrast black cast iron, cream-colored wheel hubs, and red brackets.

Combine realistic proportions with mild exaggeration in a collectible-model style. Give the chef an expressive face and natural clothing folds. Construct the machinery from clear cylinders, discs, rectangular plates, straight rods, and a few regular curved tubes, with legible part boundaries and plausible connections. Preserve the contact relationships among the character, seat, handlebars, and foot supports. Show the complete subject centered in a slightly elevated front three-quarter view against a white background, with soft studio lighting, contact shadows, clear focus, and refined material appearance. Exclude text, numbers, logos, watermarks, complex scenery, motion blur, collages, and exploded-part presentation.

Result

The delivered asset is 76-v2-final.glb, a model/gltf-binary file containing 51,757,192 bytes. The supplied final render shows the GLB reloaded as a complete static display assembly, with the chef, pan body, red frame, three wheels, controls, seat, and rear mechanical details visible. Surfaces not established by the references were inferred during modeling, and no animation, topology, UV, dimensional, printability, or runtime-performance claims are made for this case.