STEALTHMARK · HEAD TO HEAD

All comparisons ↗

Claude Haiku 5.5 vs Claude Opus 5.5

Claude Opus 5.5 scores 23.2 points higher overall (69.7 vs 46.5). Claude Opus 5.5 scores higher on Newton's Mirror, with its largest category advantage in reflections.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

46.5/100
69.7/100

General Intelligence

40.7/100
66.2/100

Visual Intelligence

53.3/100
73.9/100

Newton's Mirror

Same prompt

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

Claude Haiku 5.540.0/100

Claude Haiku 5.5: Newton's Mirror SVG benchmark
Researcher observations

The centered sphere, compact airborne black blob and differentiated cyan/crimson interaction support substantial state and placement credit. The few recognizable sphere and panel images establish limited optical relationships, but most of the required mirror system is missing. Rounded volume and localized paint warrant substantial material credit despite the soft, illustrative finish.

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.

Claude Opus 5.555.7/100

Claude Opus 5.5: Newton's Mirror SVG benchmark
Researcher observations

Open lateral sightlines, first reflected silhouettes and near-field planar panel placement are convincing. Missing outer chains, truncated repetitions and absent nested spherical room and emitter imagery remain the main weaknesses.

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
CategoryClaude Haiku 5.5Claude Opus 5.5
State at 600 ms23.5/2525.0/25
Camera and placement8.0/1515.0/15
Reflections24.5/12046.5/120
Materials and scale7.5/108.0/10
Floor-damage reward0.0/100.0/10

Selfie

Same prompt

Claude Haiku 5.546.4/100

Claude Haiku 5.5: Selfie SVG benchmark
Researcher observations

A coherent coral squid emblem with a compact mantle, restrained face and clean centered presentation. Its radial silhouette and floating placement work, but sharp arm tips, crowded roots and sparse surface modeling limit softness, construction and finish.

Simplified prompt

Draw a calm coral-red squid with eight thick arms spreading around it like flower petals. Keep the shape balanced and simple, with soft shading and a friendly face.

The exact prompt is held back to prevent benchmaxxing.

Claude Opus 5.577.6/100

Claude Opus 5.5: Selfie SVG benchmark
Researcher observations

The centered coral squid has a friendly, restrained expression and a balanced radial silhouette within a cream arch. Smooth curves and soft shading give it an appealing matte finish. The mantle is more elongated than compact, the upper arms are visually ambiguous, and the two hanging tentacles read somewhat like legs. Clean execution supports a strong stylized illustration.

Simplified prompt

Draw a calm coral-red squid with eight thick arms spreading around it like flower petals. Keep the shape balanced and simple, with soft shading and a friendly face.

The exact prompt is held back to prevent benchmaxxing.

Compare category scores
CategoryClaude Haiku 5.5Claude Opus 5.5
Prompt & likeness6.5/108.0/10
Anatomy & construction4.0/107.5/10
Space & placement6.5/108.0/10
Light & reflections4.0/107.5/10
Purposeful detail3.5/107.5/10
Artistry & character3.0/108.0/10
Materials & effects4.5/107.5/10
Visual integrity8.0/108.5/10

Mona Lisa

Different prompts

Claude Haiku 5.538.6/100

Claude Haiku 5.5: Mona Lisa SVG benchmark
Researcher observations

A recognizable but basic Mona Lisa reinterpretation. Clean contours and smooth facial lighting provide coherence and volume. Replication remains limited by the frontal pose, elongated neck, schematic expression, unresolved crossed hands and sparse landscape. Lighting is more successful than anatomical construction and purposeful detail.

Exact prompt

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

Claude Opus 5.562.5/100

Claude Opus 5.5: Mona Lisa SVG benchmark
Researcher observations

The portrait preserves the recognizable composition, dark veil, folded hands and distant landscape. Replication is limited by the elongated face and neck, conspicuous eyes and simplified smile. Hands remain connected but their rounded fingers and wrists lack convincing anatomy. Restrained colors create cohesion; shallow modeling and pervasive mottling weaken volume and painterly subtlety.

Exact prompt

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

Compare category scores
CategoryClaude Haiku 5.5Claude Opus 5.5
Prompt & likeness3.0/106.5/10
Anatomy & construction4.0/105.5/10
Space & placement4.0/106.5/10
Light & reflections5.5/106.5/10
Purposeful detail2.5/106.5/10
Artistry & character3.5/106.0/10
Materials & effects2.5/107.0/10
Visual integrity5.5/107.0/10

Beach

Same prompt

Claude Haiku 5.571.9/100

Claude Haiku 5.5: Beach SVG benchmark
Researcher observations

A cheerful, coherent beach illustration with attractive sunset colors, readable accessories and clean silhouettes. Its sun and sailboat reflections establish useful physical relationships. Simple animal construction, flattened foreground placement, sparse directional shadows and minimally differentiated surfaces limit its drawing refinement.

Exact prompt

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

Claude Opus 5.577.0/100

Claude Opus 5.5: Beach SVG benchmark
Researcher observations

A lively, well-composed beach illustration with rich sunset colors, recognizable animals and appealing accessories. The sun reflection establishes a strong central focus. Clean silhouettes and small details reward inspection, though simplified animal construction, uneven contact shadows and mostly flat materials limit depth and realism.

Exact prompt

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

Compare category scores
CategoryClaude Haiku 5.5Claude Opus 5.5
Prompt & likeness8.5/108.5/10
Anatomy & construction6.5/107.5/10
Space & placement7.0/107.0/10
Light & reflections7.0/107.5/10
Purposeful detail7.0/107.5/10
Artistry & character7.0/108.0/10
Materials & effects6.0/106.5/10
Visual integrity8.5/108.0/10

Ground Truth

Same prompt

Claude Haiku 5.547.4/100

Claude Haiku 5.5: 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.

Claude Opus 5.580.2/100

Claude Opus 5.5: 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

Same prompt

Claude Haiku 5.547.3/100

Claude Haiku 5.5: Raccoon SVG benchmark
Researcher observations

A tidy, appealing cartoon with recognizable animals and a pleasing sunset palette. Steering lacks visible controls, while schematic limbs and uncertain raccoon support weaken the requested interaction. Shallow vehicle construction, flat character lighting and modest displacement limit physical coherence. Clean shapes and readable identifying details remain strengths.

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

Claude Opus 5.576.0/100

Claude Opus 5.5: Raccoon SVG benchmark
Researcher observations

The masked raccoon clearly rides the dog's back, and the dog's forward paw reaches the steering grip. Expressive faces, vivid colors and sweeping wakes create an appealing illustration. Connected limbs remain schematic, as does the small steering assembly. The broad hull and flat foam weaken water contact; the glowing sun lacks a corresponding reflection or convincing cast shadows.

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
CategoryClaude Haiku 5.5Claude Opus 5.5
Prompt & likeness5.5/108.5/10
Anatomy & construction4.0/107.5/10
Space & placement5.5/107.5/10
Light & reflections4.0/106.0/10
Purposeful detail5.5/108.5/10
Artistry & character4.5/107.5/10
Materials & effects5.5/107.0/10
Visual integrity5.0/108.0/10

Pelican

Benchmax control

Outside General and Visual Intelligence. Excluded from the overall score for Claude Haiku 5.5 and Claude Opus 5.5.

Claude Haiku 5.564.0/100

Claude Haiku 5.5: Pelican SVG benchmark
Researcher observations

A cheerful, recognizable illustration with clean contours and a balanced warm palette. Aligned wheels and readable frame geometry establish the bicycle, but shoulder construction, feet and steering contact remain imprecise. Sparse drivetrain detail, flat materials and absent grounding shadows limit anatomical and spatial conviction.

Exact prompt

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

Claude Opus 5.572.5/100

Claude Opus 5.5: Pelican SVG benchmark
Researcher observations

The recognizable pelican, orderly bicycle wheels and harmonious sunset create an appealing illustration. Both legs connect to the body and reach pedals; no clear detached or inverted parts appear. Steering contact and seating support are less convincing, while the drivetrain remains schematic. Crisp outlines give a clean finish, but limited shading and flat surfaces leave the scene with modest depth.

Exact prompt

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

Compare category scores
CategoryClaude Haiku 5.5Claude Opus 5.5
Prompt & likeness8.0/108.5/10
Anatomy & construction5.5/106.5/10
Space & placement6.5/107.0/10
Light & reflections5.0/106.0/10
Purposeful detail6.0/107.5/10
Artistry & character6.5/107.5/10
Materials & effects5.0/106.0/10
Visual integrity8.0/108.5/10