STEALTHMARK · HEAD TO HEAD

All comparisons ↗

Gemma 4 26B A4B vs Step 5 Preview

Step 5 Preview leads Gemma 4 26B A4B by 22.5 points on StealthMark, 39.0 to 16.5. It leads on 7 of 9 benchmarks, including both games, and Gemma 4 26B A4B on 2. 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

16.5/100Overall rank 50/53
39.0/100Overall rank 36/53

General Intelligence

9.4/100
37.1/100

Visual Intelligence

28.9/100
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

  • Gemma 4 26B A4B

Open or local

Local modelOpen model · Apache 2.0

BENCHMARK BY BENCHMARK

Where each model leads

  • Gemma 4 26B A4B
  • Step 5 Preview
  • Other models
  • Median

DATASHEET

Specs and pricing

Step 5 Preview scores higher on StealthMark and takes 4× more context.

  • Gemma 4 26B A4B
  • Step 5 Preview
  • Other models
  • Median
PricePer 1M output tokens, input below
Gemma 4 26B A4B: $0.30$0.09 inputStep 5 Preview: No list price
Context windowTokens per request, max output below
Gemma 4 26B A4B: 256KStep 5 Preview: 1M4× larger
Output speedTokens per second, our measurement
Gemma 4 26B A4B: 95tokens/sStep 5 Preview: Not measured
SizeTotal and active parameters
Gemma 4 26B A4B: 25.2B3.8B activeStep 5 Preview: 600B27B active
Developer
Google DeepMindStepFun
Released
2 Apr 2026Not published
Origin
United StatesNot published
Input → output
Text, image → textText, image, video → text
Architecture
Mixture-of-Experts multimodal transformer with interleaved sliding-window and global attentionSparse 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 sourceGemma 4 26B A4B ↗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.

Gemma 4 26B A4B 4.1/100Rank 48/49

Survived 0 of 3
Foresight: complete playing field
Level 32 ptsTurn 9 / 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

LevelGemma 4 26B A4BStep 5 Preview
Level 1Second Thoughts
3 ptsOut of lives
22 ptsSurvived
Level 2Pressure Pot
2 ptsOut of lives
Pending
Level 3Hide or Flight
2 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.

Gemma 4 26B A4B 4.5/100Rank 43/50

Cleared 2 of 5
Latch: complete playing field
Level 53 ptsTurn 40 / 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

LevelGemma 4 26B A4BStep 5 Preview
Level 1Hold the door
3 ptsCleared1 beacon left
5.2 ptsCleared
Level 2Crossfeed
6.2 ptsFinished early
9 ptsCleared
Level 3Relay vault
15.2 ptsClearedFinished early · 3 beacons left
18.6 ptsCleared
Level 4Passing bay
11.8 ptsFinished early
15.6 ptsClearedFinished early · 3 beacons left
Level 5Second thoughts
3 ptsOut of turns
15 ptsFinished early

Newton's Mirror

Same task instructions

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

The judgment

Gemma 4 26B A4B

Separate airborne black and contact-compatible paint flattening earn partial state credit. Settlement, adhesion and differentiated paint transport remain unresolved. The enlarged, low sphere and thin lamp strip substantially depart from the required camera, while the defining mirror structure is missing. Weak material appearance and oversized black paint limit material credit. The retained base total is 18/170; the floor-damage reward is 0, giving 18/180.

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

CategoryGemma 4 26B A4BStep 5 Preview
State at 600 ms10.0/2520.0/25
Camera and placement2.0/1513.5/15
Reflections0.0/12020.0/120
Materials and scale2.5/107.0/10
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

Gemma 4 26B A4B

An appealing, readable sparkle-creature with luminous colour and expressive eyes. Simple geometric construction limits sophistication, while sharp body points, ambiguous hand contact and a minimally resolved jar weaken the requested soft, plump character and treasure-presenting pose.

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

CategoryGemma 4 26B A4BStep 5 Preview
Task & likeness5.5/106.0/10
Anatomy & construction4.5/104.5/10
Space & placement6.0/105.0/10
Light & reflections6.0/105.5/10
Purposeful detail5.5/105.5/10
Artistry & character4.0/105.0/10
Materials & effects6.0/106.0/10
Visual integrity7.0/105.0/10

Benchmark task

Gemma 4 26B A4B

A small luminous creature proudly holding an empty glass jar tests expression, transparent materials and the balance of character and prop.

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

Gemma 4 26B A4B

The centered dark-haired portrait provides a loose reference cue, but its elongated, nearly featureless face and undifferentiated torso fall far short of replication. Crossed hands, the subtle smile, clothing structure and landscape are unresolved. Clean silhouettes and soft gradients provide limited coherence.

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

CategoryGemma 4 26B A4BStep 5 Preview
Task & likeness0.5/104.0/10
Anatomy & construction2.0/104.0/10
Space & placement1.0/104.0/10
Light & reflections2.0/103.5/10
Purposeful detail1.0/104.0/10
Artistry & character2.0/104.0/10
Materials & effects1.0/103.0/10
Visual integrity4.5/103.5/10

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

Gemma 4 26B A4B

A clean, balanced sunset scene demonstrates the beach, birds and a centered reflection. Animals and play receive only minimal treatment. Coherent layers and controlled gradients support readability, but unresolved aquatic anatomy, sparse details and weak surface modeling keep the drawing basic.

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

CategoryGemma 4 26B A4BStep 5 Preview
Task & likeness3.5/104.0/10
Anatomy & construction2.0/103.0/10
Space & placement3.5/103.5/10
Light & reflections5.0/103.5/10
Purposeful detail1.0/103.0/10
Artistry & character1.5/103.0/10
Materials & effects2.0/103.5/10
Visual integrity6.0/104.0/10

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

Gemma 4 26B A4B

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

Gemma 4 26B A4B

The animal stack is recognizable, and clean silhouettes support a basic illustration. Steering is not depicted, while the craft's construction and water contact remain unresolved. Sparse anatomical and material detail limit the finish. The sky gradient and glow provide atmospheric lighting, but the subjects lack corresponding modeling.

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

CategoryGemma 4 26B A4BStep 5 Preview
Task & likeness2.5/104.0/10
Anatomy & construction2.5/104.5/10
Space & placement4.0/103.0/10
Light & reflections3.5/103.5/10
Purposeful detail2.5/104.0/10
Artistry & character3.0/103.5/10
Materials & effects2.0/103.0/10
Visual integrity7.0/103.0/10

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

Gemma 4 26B A4B

A clean, balanced icon suggests a pelican riding a bicycle. Continuous contours and restrained colors support its finish, but the abbreviated bill, undifferentiated body, unresolved riding contacts and incomplete bicycle mechanism limit likeness and construction. Sparse detail and flat lighting keep the drawing basic despite its clarity.

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

CategoryGemma 4 26B A4BStep 5 Preview
Task & likeness3.0/103.0/10
Anatomy & construction3.0/103.0/10
Space & placement5.0/103.0/10
Light & reflections3.0/103.0/10
Purposeful detail3.0/104.0/10
Artistry & character2.5/103.0/10
Materials & effects2.5/103.0/10
Visual integrity5.0/102.0/10

Benchmark task

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