STEALTHMARK · HEAD TO HEAD

All comparisons ↗

GPT-6 Astra vs GLM-5.3-Flash (Ox Alpha)

GPT-6 Astra scores 16.8 points higher overall (70.3 vs 53.5). GPT-6 Astra scores higher on Newton's Mirror, with its largest category advantage in reflections.

See the difference ↘
All results ↗
Model guide ↗

Overall StealthMark

70.3/100
53.5/100

General Intelligence

64.0/100
48.7/100

Visual Intelligence

77.9/100
60.3/100

Newton's Mirror

Same prompt

Physics, perspective and recursive reflections, tested together. Explore the test ↗

GPT-6 Astra62.7/100

GPT-6 Astra: Newton's Mirror SVG benchmark
Researcher observations

Major placement and the visible contact sequence are convincing. Dense planar light repetition, recognizable spherical panel images and simplified nested sphere motifs provide useful optical structure. Missing outer chains, truncated distant sequences and incorrect bead-like belts still leave the recursive reflections substantially incomplete.

Simplified prompt

A shiny steel ball spins in a room with mirrored walls and a mirrored floor. Red, cyan and black paint float above it. Gravity suddenly switches on. Draw the scene a moment later, as the ball drops and paint begins landing on its spinning surface. Show the reflections repeating through the room.

The exact prompt is held back to prevent benchmaxxing.

GLM-5.3-Flash (Ox Alpha)27.4/100

Ox Alpha: Newton's Mirror SVG benchmark
Researcher observations

The setup is recognizable, but the paint interaction remains poorly established and off-center. The first wall and floor images provide coherent placement, scale, silhouette and occlusion cues, with faint lower echoes. Missing receding chains, nested imagery and ordered emitter mappings leave the reflection system heavily simplified. Dull shading does not convincingly depict polished steel or mirrors.

Simplified prompt

A shiny steel ball spins in a room with mirrored walls and a mirrored floor. Red, cyan and black paint float above it. Gravity suddenly switches on. Draw the scene a moment later, as the ball drops and paint begins landing on its spinning surface. Show the reflections repeating through the room.

The exact prompt is held back to prevent benchmaxxing.

Compare category scores
CategoryGPT-6 AstraGLM-5.3-Flash (Ox Alpha)
State at 600 ms25.0/2512.5/25
Camera and placement15.0/1510.0/15
Reflections54.0/12021.5/120
Materials and scale10.0/103.5/10
Floor-damage reward0.0/100.0/10

Selfie

Different prompts

GPT-6 Astra77.0/100

GPT-6 Astra: Selfie SVG benchmark
Researcher observations

The sixfold silhouette, pale coloring, outward heads and central hexagon closely match the brief. Fine scales and soft shading support rounded volume. However, several crossings have conspicuous rectangular seams and weak contact shadows, making the weave feel assembled. Small heads have limited anatomical definition, and the repetitive texture conveys less organic life than requested.

Simplified prompt

Draw a pale six-headed snake woven into a symmetrical knot. Its body passes over and under itself around a small hexagonal opening, with six little heads facing outward. Add lifelike scales, curved muscles and shadows without losing the clear knot shape.

The exact prompt is held back to prevent benchmaxxing.

GLM-5.3-Flash (Ox Alpha)77.5/100

Ox Alpha: Selfie SVG benchmark
Researcher observations

The expressive emerald face, ivory horns, gold accents and fine veining closely match the brief. Rounded volumes and layered shading create an appealing mascot. Rear hoof placement is cramped, lighting is somewhat inconsistent, and conspicuous white patches behind the muzzle interrupt the finish. Surfaces remain visibly stylized rather than photographic.

Simplified prompt

Draw a friendly green bull with a big head, ivory horns and gold details. It lifts one front hoof and wears a gold alpha symbol on its forehead. Make its rounded body look like polished stone.

The exact prompt is held back to prevent benchmaxxing.

Compare category scores
CategoryGPT-6 AstraGLM-5.3-Flash (Ox Alpha)
Prompt & likeness8.5/108.5/10
Anatomy & construction8.0/107.5/10
Space & placement7.5/107.5/10
Light & reflections7.0/108.0/10
Purposeful detail7.5/108.0/10
Artistry & character8.0/107.5/10
Materials & effects7.0/108.0/10
Visual integrity7.0/107.0/10

Mona Lisa

Different prompts

GPT-6 Astra68.2/100

GPT-6 Astra: Mona Lisa SVG benchmark
Researcher observations

The crossed hands, restrained smile, dark clothing and hazy landscape clearly recall the Mona Lisa. Replication is limited by the frontal, elongated face, schematic eyes and stiff fingers. Soft shading gives basic volume, but broad garment shapes and angular scenery lack the original’s nuanced depth and painterly transitions. The frame and composition are controlled; surface grain adds texture without resolving the simplified drawing.

Exact prompt

Replicate mona-lisa as SVG Make it a masterpiece, no tools allowed

GLM-5.3-Flash (Ox Alpha)49.5/100

Ox Alpha: Mona Lisa SVG benchmark
Researcher observations

The crossed arms, green dress and hazy landscape clearly evoke Mona Lisa. Replication is limited by the frontal pose, oversized eyes, elongated face and pronounced lips. Overlapping finger strips lack convincing knuckles and palm volume. The balanced framing and soft shading provide cohesion, but simplified anatomy, repetitive folds and blurred scenery keep the portrait basic.

Exact prompt

Generate an SVG of mona-lisa Make it a masterpiece You may NOT use tools, browser, image analysis. Directly output the SVG as code

Compare category scores
CategoryGPT-6 AstraGLM-5.3-Flash (Ox Alpha)
Prompt & likeness7.0/104.5/10
Anatomy & construction6.5/104.5/10
Space & placement6.5/106.0/10
Light & reflections7.0/106.0/10
Purposeful detail6.5/105.0/10
Artistry & character7.0/105.0/10
Materials & effects7.0/105.5/10
Visual integrity7.5/104.0/10

Beach

Different prompts

GPT-6 Astra81.8/100

GPT-6 Astra: Beach SVG benchmark
Researcher observations

A richly composed sunset beach illustration with cohesive peach, mauve and teal colors. Sweeping palm fronds, a scalloped umbrella and curved shorelines frame the sun and its aligned reflection. Distinct animal silhouettes and carefully distributed beach details reward inspection. Simplified human and dog anatomy, slight dolphin splash cues and inconsistent equipment shadows limit physical conviction.

Exact prompt

Create an SVG of a amazing beach scene. Sunset, animals, fun, birds, reflections Make it a masterpiece, no tools allowed

GLM-5.3-Flash (Ox Alpha)77.7/100

Ox Alpha: Beach SVG benchmark
Researcher observations

Framing palms, a vivid sunset palette and balanced kite and umbrella create an appealing beach illustration. Layered water and shoreline give the scene depth, while aligned reflections reinforce the sunset. Small accessories add character and reward inspection. Figures and animals remain schematic, some overlaps are difficult to read, and simplified shadows and flat surfaces limit physical richness.

Exact prompt

Create an SVG of a amazing beach scene. Sunset, animals, fun, birds, reflections Make it a masterpiece You may NOT use tools, browser, image analysis. Directly output the SVG as code

Compare category scores
CategoryGPT-6 AstraGLM-5.3-Flash (Ox Alpha)
Prompt & likeness9.0/108.0/10
Anatomy & construction7.5/107.0/10
Space & placement8.5/108.0/10
Light & reflections7.5/108.0/10
Purposeful detail8.5/108.0/10
Artistry & character9.0/108.5/10
Materials & effects7.5/106.5/10
Visual integrity8.0/108.0/10

Ground Truth

Same prompt

GPT-6 Astra68.7/100

GPT-6 Astra: Ground Truth SVG benchmark
Simplified prompt

A dark steel room has four pillars and two warm ceiling lights. A wooden plank is placed over a computer mouse, with a blueberry on the plank above it. Draw the scene a few seconds later, showing how the objects settle and how the light, shadows and surfaces look.

The exact prompt is held back to prevent benchmaxxing.

GLM-5.3-Flash (Ox Alpha)64.3/100

Ox Alpha: Ground Truth SVG benchmark
Simplified prompt

A dark steel room has four pillars and two warm ceiling lights. A wooden plank is placed over a computer mouse, with a blueberry on the plank above it. Draw the scene a few seconds later, showing how the objects settle and how the light, shadows and surfaces look.

The exact prompt is held back to prevent benchmaxxing.

Raccoon

GPT-6 Astra78.8/100

GPT-6 Astra: Raccoon SVG benchmark
Researcher observations

Expressive animals, a balanced sunset composition and sweeping wake make an appealing illustration. The raccoon rides the dog's back, and the dog's connected foreleg reaches the controls. The compact steering assembly and layered paws are less clearly constructed. Lighting remains mostly flat, with little shadow or sun reflection. Clean contours and purposeful details support strong illustrative execution.

Exact prompt

Generate an SVG: A raccoon sits on a dog's back while the dog steers a jet ski Make it a masterpiece, no tools allowed

Compare category scores
CategoryGPT-6 AstraGLM-5.3-Flash (Ox Alpha)
Prompt & likeness8.5/10—
Anatomy & construction7.5/10—
Space & placement8.0/10—
Light & reflections6.5/10—
Purposeful detail8.0/10—
Artistry & character8.5/10—
Materials & effects8.0/10—
Visual integrity8.5/10—

Pelican

Benchmax control

Outside General and Visual Intelligence. Excluded from the overall score for GPT-6 Astra and GLM-5.3-Flash (Ox Alpha).

GPT-6 Astra75.8/100

GPT-6 Astra: Pelican SVG benchmark
Researcher observations

The pelican's long bill, flowing scarf and warm coastal palette create a charming, balanced illustration. Wheels, spokes and frame are largely coherent, but neither wing reaches the handlebars, and the rear leg's pedal contact is unclear. Flat lighting and simplified surfaces limit depth despite clean outlines and attractive feather detail.

Exact prompt

Generate an SVG of a pelican riding a bicycle Make it a masterpiece, no tools allowed

GLM-5.3-Flash (Ox Alpha)74.8/100

Ox Alpha: Pelican SVG benchmark
Researcher observations

The pelican, bicycle and seaside setting form a clear, appealing illustration. Circular wheels, readable frame tubes and clean silhouettes support the action. The legs connect to the body, but the far foot's pedal contact and wing-to-handlebar junction are cramped. Simple shading, uniform wave marks and limited surface texture keep the rendering relatively flat.

Exact prompt

Generate a still SVG of a pelican riding a bicycle Make it a masterpiece You may NOT use tools, browser, image analysis. Directly output the SVG as code

Compare category scores
CategoryGPT-6 AstraGLM-5.3-Flash (Ox Alpha)
Prompt & likeness8.0/108.5/10
Anatomy & construction6.5/107.0/10
Space & placement7.5/107.5/10
Light & reflections7.0/106.5/10
Purposeful detail8.0/107.0/10
Artistry & character8.5/108.0/10
Materials & effects7.5/106.5/10
Visual integrity8.0/108.0/10