STEALTHMARK · HEAD TO HEAD

All comparisons ↗

MAI-Thinking-1 vs Step 5 Preview

Step 5 Preview leads MAI-Thinking-1 by 24.9 points on StealthMark, 39.0 to 14.1. It leads on 7 of 9 benchmarks, including both games, and MAI-Thinking-1 on 2. Step 5 Preview scores higher on Newton's Mirror, with its largest category advantage in state at 600 ms.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

14.1/100Overall rank 51/53
39.0/100Overall rank 36/53

General Intelligence

7.4/100
37.1/100

Visual Intelligence

24.6/100
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

  • MAI-Thinking-1Single-step reasoning not measured

BENCHMARK BY BENCHMARK

Where each model leads

  • MAI-Thinking-1
  • Step 5 Preview
  • Other models
  • Median

DATASHEET

Specs and pricing

Step 5 Preview scores higher on StealthMark and takes 4.1× more context.

  • MAI-Thinking-1
  • Step 5 Preview
  • Other models
  • Median
PricePer 1M output tokens, input below
MAI-Thinking-1: $8$2 inputStep 5 Preview: No list price
Context windowTokens per request, max output below
MAI-Thinking-1: 256K64K max outputEstimatedStep 5 Preview: 1M4.1× larger
Output speedTokens per second, our measurement
MAI-Thinking-1: 93tokens/sStep 5 Preview: Not measured
SizeTotal and active parameters
MAI-Thinking-1: 1T35B activeEstimatedStep 5 Preview: 600B27B active
Developer
Microsoft AIStepFun
Released
12 Aug 2026Not published
Origin
United StatesNot published
Input → output
Text → textText, image, video → text
Architecture
Sparse Mixture-of-Experts (MoE)Sparse 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 sourceMAI-Thinking-1 ↗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.

MAI-Thinking-1 6.4/100Rank 46/49

Survived 0 of 3
Foresight: complete playing field
Level 31 ptsTurn 8 / 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

LevelMAI-Thinking-1Step 5 Preview
Level 1Second Thoughts
3 ptsOut of lives
22 ptsSurvived
Level 2Pressure Pot
3 ptsOut of lives
Pending
Level 3Hide or Flight
1 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.

MAI-Thinking-1 3.6/100Rank 46/50

Cleared 1 of 5
Latch: complete playing field
Level 511.8 ptsTurn 16 / 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

LevelMAI-Thinking-1Step 5 Preview
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
2 ptsOut of turns
9 ptsCleared
Level 3Relay vault
2.2 ptsFinished early
18.6 ptsCleared
Level 4Passing bay
3 ptsFinished early
15.6 ptsClearedFinished early · 3 beacons left
Level 5Second thoughts
11.8 ptsFinished early
15 ptsFinished early

Newton's Mirror

Same task instructions

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

The judgment

MAI-Thinking-1

The small paint features and convex shading deserve moderate material credit. The sphere remains undersized and low, with black unusually close to its apex. Crimson and cyan do not establish the required impact histories. Faint reflected echoes supply limited optical evidence, while the defining recursive room, nested sphere and emitter structures remain unestablished.

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

CategoryMAI-Thinking-1Step 5 Preview
State at 600 ms3.0/2520.0/25
Camera and placement2.5/1513.5/15
Reflections4.0/12020.0/120
Materials and scale2.0/107.0/10
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

MAI-Thinking-1

A recognizable lantern-and-moth illustration with appealing coloured wings and tidy outer contours. Simple construction, a small focal subject and crowded overlaps limit character and depth. The face, six-leg anatomy, seedling and antenna-driven watering joke remain insufficiently readable.

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

CategoryMAI-Thinking-1Step 5 Preview
Task & likeness4.0/106.0/10
Anatomy & construction4.5/104.5/10
Space & placement5.0/105.0/10
Light & reflections4.5/105.5/10
Purposeful detail4.0/105.5/10
Artistry & character5.0/105.0/10
Materials & effects5.0/106.0/10
Visual integrity6.0/105.0/10

Benchmark task

MAI-Thinking-1

A fluffy moth tends a lantern greenhouse amid a gardening mishap. The scene tests insect anatomy, expressive action, translucent wings and the interaction of light with small objects.

Step 5 Preview

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

Mona Lisa

Different task instructions

Precise visual recall and portrait reconstruction.

The judgment

MAI-Thinking-1

A recognizable Mona Lisa motif emerges from dark hair, a pale face and a landscape band. Replication remains weak: the oversized frontal head, broad smile, unresolved torso and missing crossed-hand anatomy diverge substantially from the reference. Simple fills and sparse strokes provide limited volume, material definition or painterly complexity.

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

CategoryMAI-Thinking-1Step 5 Preview
Task & likeness1.0/104.0/10
Anatomy & construction0.5/104.0/10
Space & placement1.0/104.0/10
Light & reflections1.5/103.5/10
Purposeful detail1.0/104.0/10
Artistry & character1.5/104.0/10
Materials & effects0.5/103.0/10
Visual integrity3.0/103.5/10

Benchmark task

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

Beach

Different task instructions

Creativity, perspective and visual beauty.

The judgment

MAI-Thinking-1

A coherent, cheerful beach scene with balanced placement, clean shapes and controlled sunset colours. Animals, birds and playful accessories are recognizable, but basic dolphin construction, weak grounding, schematic reflections and sparse material detail limit the drawing's accomplishment.

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

CategoryMAI-Thinking-1Step 5 Preview
Task & likeness6.0/104.0/10
Anatomy & construction4.0/103.0/10
Space & placement5.0/103.5/10
Light & reflections4.0/103.5/10
Purposeful detail4.5/103.0/10
Artistry & character4.0/103.0/10
Materials & effects3.5/103.5/10
Visual integrity4.5/104.0/10

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

MAI-Thinking-1

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

Different task instructions

Scene composition and interacting subjects.

The judgment

MAI-Thinking-1

The recognizable raccoon and clean, shaded forms make a readable cartoon. The dog's steering remains undemonstrated, while ambiguous seating contacts and a minimally constructed craft weaken the central interaction. Controlled gradients and edges deserve credit, but anatomy, purposeful detail and water displacement remain basic.

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

CategoryMAI-Thinking-1Step 5 Preview
Task & likeness3.0/104.0/10
Anatomy & construction2.0/104.5/10
Space & placement5.0/103.0/10
Light & reflections3.5/103.5/10
Purposeful detail3.0/104.0/10
Artistry & character2.5/103.5/10
Materials & effects3.5/103.0/10
Visual integrity5.5/103.0/10

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

Different task instructions

Subject construction and spatial relationships.

The judgment

MAI-Thinking-1

A recognizable pelican and bicycle combine clean rounded forms, expressive facial details and effective surface shading. Aligned wheels and soft shadows establish placement, but unresolved pedal engagement, absent steering contact and sparse drivetrain construction weaken the riding action. Limited feather complexity and shallow space constrain an otherwise appealing illustration.

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

CategoryMAI-Thinking-1Step 5 Preview
Task & likeness7.0/103.0/10
Anatomy & construction4.5/103.0/10
Space & placement5.5/103.0/10
Light & reflections7.0/103.0/10
Purposeful detail5.0/104.0/10
Artistry & character5.0/103.0/10
Materials & effects4.5/103.0/10
Visual integrity7.0/102.0/10

Benchmark task

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