STEALTHMARK · HEAD TO HEAD

All comparisons ↗

GPT-5.6 Terra vs Step 5 Preview

GPT-5.6 Terra leads Step 5 Preview by 6.8 points on StealthMark, 45.8 to 39.0. It leads on 8 of 9 benchmarks, including both games, and Step 5 Preview on 1. Step 5 Preview scores higher on Newton's Mirror, with its largest category advantage in reflections.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

45.8/100Overall rank 25/53
39.0/100Overall rank 36/53

General Intelligence

41.9/100
37.1/100

Visual Intelligence

52.7/100
41.9/100

Reasoning efficiency

On every task, result quality and reasoning tokens are each compared with all measured models and weighed equally. Each dot is one model.

Single-step reasoning Result quality weighed against reasoning tokens on a single problem

Multi-step reasoning Result quality weighed against reasoning tokens across connected steps, as in agentic workflows

  • GPT-5.6 Terra

BENCHMARK BY BENCHMARK

Where each model leads

  • GPT-5.6 Terra
  • Step 5 Preview
  • Other models
  • Median

DATASHEET

Specs and pricing

Both take about 1.05M tokens of context.

  • GPT-5.6 Terra
  • Step 5 Preview
  • Other models
  • Median
PricePer 1M output tokens, input below
GPT-5.6 Terra: $12$2 inputStep 5 Preview: No list price
Context windowTokens per request, max output below
GPT-5.6 Terra: 1.05MTied128K max outputStep 5 Preview: 1MTied
Output speedTokens per second, our measurement
GPT-5.6 Terra: 102tokens/sStep 5 Preview: Not measured
SizeTotal and active parameters
GPT-5.6 Terra: Not disclosedStep 5 Preview: 600B27B active
Developer
OpenAIStepFun
Released
9 Jul 2026Not published
Origin
United StatesNot published
Input → output
Text, image → textText, image, video → text
Architecture
Not publishedSparse mixture-of-experts; 92 layers (88 MoE), 352 routed experts with 8 selected per token plus a shared expert; grouped-query attention with 512-token sliding windows and full attention every fourth layer; 3 multi-token-prediction layers; vision tower for image and video input

Full datasheets, every price and sourceGPT-5.6 Terra ↗Step 5 Preview ↗

Anticipating threats and planning ahead. Collect coins while avoiding ghosts. Ghosts move before the player each turn, and collisions cost a life. 3 levels in sequence, up to 66 turns in total. Earlier moves and outcomes carry forward, testing learning across levels.

GPT-5.6 Terra 30.1/100Rank 31/49

Survived 0 of 3
Foresight: complete playing field
Level 36 ptsTurn 7 / 20

Step 5 Preview 59.9/100Rank 42/491 of 3 levels

1 of 3 played
Foresight: complete playing field
Level 122 ptsTurn 28 / 28

Points per level

LevelGPT-5.6 TerraStep 5 Preview
Level 1Second Thoughts
5 ptsOut of lives
22 ptsSurvived
Level 2Pressure Pot
9 ptsOut of lives
Pending
Level 3Hide or Flight
6 ptsOut of lives
Pending

Cooperative planning and learning from experience. Coordinate two robots to push numbered cores onto matching sockets. Pressure plates hold doors open; moving either robot uses one turn. 5 levels in sequence, up to 152 turns in total. Earlier moves and outcomes carry forward, testing learning across levels.

GPT-5.6 Terra 80.0/100Rank 10/50

Cleared 5 of 5
Latch: complete playing field
Level 520.6 ptsTurn 27 / 40

Step 5 Preview 40.7/100Rank 28/50

Cleared 4 of 5
Latch: complete playing field
Level 515 ptsTurn 35 / 40

Points per level

LevelGPT-5.6 TerraStep 5 Preview
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
9.6 ptsCleared
9 ptsCleared
Level 3Relay vault
19.2 ptsCleared
18.6 ptsCleared
Level 4Passing bay
18.2 ptsClearedFinished early · 3 beacons left
15.6 ptsClearedFinished early · 3 beacons left
Level 5Second thoughts
20.6 ptsCleared
15 ptsFinished early

Newton's Mirror

Same task instructions

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

The judgment

GPT-5.6 Terra

The real sphere and airborne black blob are substantially too low in the complete frame after accounting for aspect ratio. The first wall and floor reflections have incompatible placement, relative size or silhouette. Both lateral regions lack the references' receding lane-and-chain organization. The few recognizable mirror images provide limited optical structure. The main, side and floor spheres replace nested mirror imagery with smooth bands, while sparse planar light patches and broad spherical highlights fail to establish coherent repeated emitter mapping.

Step 5 Preview

The main sphere, airborne black and small colored features occupy broadly appropriate positions. Most of the defining mirror system is absent, and paint interaction remains only partially supported. The small paint features do not visibly establish a separate volume-scale violation.

Score breakdown

CategoryGPT-5.6 TerraStep 5 Preview
State at 600 ms24.0/2520.0/25
Camera and placement6.0/1513.5/15
Reflections7.0/12020.0/120
Materials and scale5.5/107.0/10
Floor-damage reward0.0/100.0/10

Benchmark task

A spinning metal sphere and paint in a mirrored room test physical reasoning, perspective and consistency across repeated reflections.

Selfie

Different task instructions

Visual construction and creative interpretation.

The judgment

GPT-5.6 Terra

Six ivory heads and rounded lobes create a striking radial emblem around a clear hexagonal opening. Soft shading gives the bodies volume, but abrupt rectangular and diagonal band endings make the continuous alternating weave ambiguous. Repetitive scale lines and simplified faces convey an ornamental serpent more convincingly than living anatomy.

Step 5 Preview

A charming diagonal climber and warm palette communicate the central idea. Simple rounded construction, crowded limb counting, uncertain footholds and an unattached upper shelf limit its physical coherence. Growth bands and the yarn trail add specificity, but the requested material distinctions and unravelled bootie remain underdeveloped.

Score breakdown

CategoryGPT-5.6 TerraStep 5 Preview
Task & likeness7.5/106.0/10
Anatomy & construction5.5/104.5/10
Space & placement7.0/105.0/10
Light & reflections7.5/105.5/10
Purposeful detail6.5/105.5/10
Artistry & character5.5/105.0/10
Materials & effects6.5/106.0/10
Visual integrity5.0/105.0/10

Benchmark task

GPT-5.6 Terra

A pale serpent woven into an interlocking knot tests connected anatomy, repeated structure and readable overlaps. Organic surface detail must work with the overall geometry.

Step 5 Preview

Character scenes test expression, anatomy, materials and coherent visual construction.

Mona Lisa

Same task instructions

Precise visual recall and portrait reconstruction.

The judgment

GPT-5.6 Terra

The dark clothing, central portrait and restrained smile evoke the Mona Lisa, but replication is limited by the broad face, elongated neck, nearly frontal pose and indistinct landscape. Soft gradients give the face some volume. Crossed arms become rounded masses without readable fingers or wrists, and the expression lacks the original's subtlety. Clean framing and curves support an otherwise basic, uneven portrait.

Step 5 Preview

A recognizable, coherent simplification with purposeful facial contours and differentiated fingers. These provide more construction and detail than the broad silhouette alone suggests. The oversized face, frontal seated pose, unresolved wrists and sparse landscape still substantially limit likeness, depth and finish.

Score breakdown

CategoryGPT-5.6 TerraStep 5 Preview
Task & likeness4.0/104.0/10
Anatomy & construction3.0/104.0/10
Space & placement4.5/104.0/10
Light & reflections5.0/103.5/10
Purposeful detail4.0/104.0/10
Artistry & character4.0/104.0/10
Materials & effects4.0/103.0/10
Visual integrity6.0/103.5/10

Benchmark task

An SVG interpretation of the Mona Lisa tests facial likeness, anatomy and the composition of a familiar portrait.

Beach

Same task instructions

Creativity, perspective and visual beauty.

The judgment

GPT-5.6 Terra

Warm sunset gradients, framing palms and colorful beach activities create an appealing scene. Animals, birds and a sun reflection fulfill the brief. The dog and turtle remain schematic, foreground depth is shallow, and the reflection forms rigid bands. Clean silhouettes and readable accessories support a competent illustration, but limited anatomy, material variation and lighting keep it below exceptional.

Step 5 Preview

A recognizable sunset beach with clear warm-cool contrast and playful accessories. Basic animal construction, shallow foreground placement and schematic reflections limit spatial and physical credibility. Sparse descriptive detail and abruptly ended sky bands constrain the finish.

Score breakdown

CategoryGPT-5.6 TerraStep 5 Preview
Task & likeness7.5/104.0/10
Anatomy & construction5.0/103.0/10
Space & placement6.0/103.5/10
Light & reflections5.0/103.5/10
Purposeful detail6.5/103.0/10
Artistry & character5.0/103.0/10
Materials & effects5.5/103.5/10
Visual integrity5.0/104.0/10

Benchmark task

A busy shoreline at sunset tests scene composition, animal forms and the relationship between light and water reflections.

Ground Truth

Same task instructions

Physical reasoning, light and visual reconstruction.

The judgment

GPT-5.6 Terra

This benchmark has not been assessed yet.

Step 5 Preview

This benchmark has not been assessed yet.

Benchmark task

A plank, computer mouse and blueberry in a steel room test physical stability, spatial relationships, materials and lighting.

Raccoon

Same task instructions

Scene composition and interacting subjects.

The judgment

GPT-5.6 Terra

Expressive faces, a recognizable raccoon perched on the dog, and a sweeping wake make an appealing cartoon. The principal weakness is the steering interaction: the dog's paw sits beside the control assembly without a convincing handlebar grip. Bold outlines and coherent layering provide polish, while simplified shading, compressed seating geometry and angular splashes limit depth and physical credibility.

Step 5 Preview

Recognizable animals, a bright palette and identifying accessories support a basic cartoon. Primitive anatomy, partial seated support, unresolved steering and weak hull-water contact limit the central action. Lighting and spray provide atmosphere without establishing convincing physical depth.

Score breakdown

CategoryGPT-5.6 TerraStep 5 Preview
Task & likeness7.5/104.0/10
Anatomy & construction5.0/104.5/10
Space & placement7.0/103.0/10
Light & reflections6.0/103.5/10
Purposeful detail7.0/104.0/10
Artistry & character5.0/103.5/10
Materials & effects7.0/103.0/10
Visual integrity7.5/103.0/10

Benchmark task

A jet-ski scene pairs a canine driver with a raccoon passenger. It tests the placement and contact of interacting subjects, along with vehicle construction.

Pelican

Same task instructions

Subject construction and spatial relationships.

The judgment

GPT-5.6 Terra

The oversized bill, expressive eye and balanced twilight palette make the cyclist immediately readable. Clean contours and wheel details support the illustration, but the rabbit-like head tufts weaken pelican likeness. The wing does not clearly grip the handlebar, pedal contact is uneven, and the drivetrain is incomplete. Lighting remains mostly decorative.

Step 5 Preview

The pelican and bicycle are recognizable, with a pleasant palette and basic diffuse shading. The riding relationship fails physically: the bird is unsupported, the front wheel is unmounted, and pedalling is unresolved. Simple geometric construction and sparse material treatment further limit the drawing.

Score breakdown

CategoryGPT-5.6 TerraStep 5 Preview
Task & likeness6.0/103.0/10
Anatomy & construction6.0/103.0/10
Space & placement7.0/103.0/10
Light & reflections6.5/103.0/10
Purposeful detail7.0/104.0/10
Artistry & character6.5/103.0/10
Materials & effects6.5/103.0/10
Visual integrity7.0/102.0/10

Benchmark task

A cycling pelican tests anatomy, riding posture and contact between the bird and the bicycle.