STEALTHMARK · HEAD TO HEAD

All comparisons ↗

Qwen3.8-27B vs Step 5 Preview

Qwen3.8-27B leads Step 5 Preview by 12.7 points on StealthMark, 51.7 to 39.0. It leads on 8 of 9 benchmarks, including Foresight, and Step 5 Preview on 1, including Latch. Qwen3.8-27B scores higher on Beach, with its largest category advantage in anatomy & construction.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

51.7/100Overall rank 14/53
39.0/100Overall rank 36/53

General Intelligence

46.2/100
37.1/100

Visual Intelligence

61.4/100
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

Open or local

Local modelOpen model · Apache-2.0

BENCHMARK BY BENCHMARK

Where each model leads

  • Qwen3.8-27B
  • Step 5 Preview
  • Other models
  • Median

DATASHEET

Specs and pricing

Both take about 1M tokens of context.

  • Qwen3.8-27B
  • Step 5 Preview
  • Other models
  • Median
PricePer 1M output tokens, input below
Qwen3.8-27B: $1.70$0.42 inputStep 5 Preview: No list price
Context windowTokens per request, max output below
Qwen3.8-27B: 1MTied128K max outputStep 5 Preview: 1MTied
Output speedTokens per second, our measurement
Qwen3.8-27B: 78tokens/sStep 5 Preview: Not measured
SizeTotal and active parameters
Qwen3.8-27B: 27BAll activeEstimatedStep 5 Preview: 600B27B active
Developer
QwenStepFun
Released
14 Aug 2026Not published
Origin
ChinaNot published
Input → output
Text, image, video → textBoth
Architecture
Dense causal language model with vision encoder; hybrid Gated DeltaNet and gated attention; MTP-trainedSparse 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 sourceQwen3.8-27B ↗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.

Qwen3.8-27B 63.0/100Rank 7/49

Survived 1 of 3
Foresight: complete playing field
Level 311 ptsTurn 16 / 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

LevelQwen3.8-27BStep 5 Preview
Level 1Second Thoughts
26 ptsSurvived
22 ptsSurvived
Level 2Pressure Pot
12 ptsOut of lives
Pending
Level 3Hide or Flight
11 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.

Qwen3.8-27B 38.6/100Rank 30/50

Cleared 2 of 5
Latch: complete playing field
Level 515.8 ptsTurn 31 / 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

LevelQwen3.8-27BStep 5 Preview
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
9.8 ptsCleared
9 ptsCleared
Level 3Relay vault
11.2 ptsFinished early
18.6 ptsCleared
Level 4Passing bay
13.2 ptsFinished early
15.6 ptsClearedFinished early · 3 beacons left
Level 5Second thoughts
15.8 ptsFinished early
15 ptsFinished early

Newton's Mirror

Same task instructions

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

The judgment

Qwen3.8-27B

The room fills the complete 16:9 frame without visible borders or labels. Central placement fits the wider frame, and the paint colors follow the expected vertical order. Bilateral side images, diminishing echoes and an overlapping floor image provide limited mirror depth. Broad sphere surfaces lack ordered nested room imagery, while planar panel sequences and outer chains remain missing or unresolved. Soft highlights do not establish emitter mapping, while paint attachment and transport history are only partly supported.

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

CategoryQwen3.8-27BStep 5 Preview
State at 600 ms23.0/2520.0/25
Camera and placement13.0/1513.5/15
Reflections24.0/12020.0/120
Materials and scale6.5/107.0/10
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

Qwen3.8-27B

A cohesive emerald-and-gold illustration with a gentle expression, flowing robes, luminous halo and three clearly ringed orbs. Limbs remain connected, and the asymmetric legs communicate floating. The rounded face reads somewhat bear-like; the raised palm faces outward rather than upward, and chest-hand fingers are crowded. Clean ornamentation exceeds the comparatively simple anatomical modeling and shading.

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

CategoryQwen3.8-27BStep 5 Preview
Task & likeness8.0/106.0/10
Anatomy & construction7.5/104.5/10
Space & placement8.0/105.0/10
Light & reflections7.5/105.5/10
Purposeful detail8.0/105.5/10
Artistry & character7.5/105.0/10
Materials & effects7.5/106.0/10
Visual integrity8.5/105.0/10

Benchmark task

Qwen3.8-27B

A floating capybara in ornate robes tests character anatomy, layered fabric and the placement of surrounding luminous forms.

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

Qwen3.8-27B

The centered portrait, dark clothing and overlapping hands evoke the Mona Lisa, but replication is weak. The face is excessively elongated, the eyes stare broadly, and the hands reduce to rounded masses with shallow finger marks. Soft shading creates some volume, while the landscape remains indistinct. The precise gold frame is more resolved than the 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

CategoryQwen3.8-27BStep 5 Preview
Task & likeness4.0/104.0/10
Anatomy & construction4.5/104.0/10
Space & placement4.5/104.0/10
Light & reflections5.0/103.5/10
Purposeful detail5.0/104.0/10
Artistry & character5.0/104.0/10
Materials & effects5.0/103.0/10
Visual integrity5.0/103.5/10

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

Qwen3.8-27B

The sweeping palm, glowing sunset and purple-to-orange palette create an appealing beach composition. Birds, animals and reflections fulfill the brief, with clean silhouettes and readable small accents. Schematic anatomy, repetitive shoreline geometry and limited perspective integration constrain depth. Foreground lighting and uniform reflection bars provide atmosphere but little physical coherence.

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

CategoryQwen3.8-27BStep 5 Preview
Task & likeness8.5/104.0/10
Anatomy & construction6.5/103.0/10
Space & placement6.0/103.5/10
Light & reflections6.0/103.5/10
Purposeful detail7.0/103.0/10
Artistry & character7.0/103.0/10
Materials & effects6.5/103.5/10
Visual integrity7.0/104.0/10

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

Qwen3.8-27B

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

Qwen3.8-27B

The raccoon clearly rides on the dog's back, and the dog reaches the jet ski's steering grip. Cheerful expressions, a bright palette and sweeping wake make the action readable. Anatomy and machinery remain simplified, with overlapping paws and limited seat definition. Flat lighting, sparse surface detail and broad white splash shapes limit depth and finish. A clipped gray fragment interrupts the upper-left corner.

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

CategoryQwen3.8-27BStep 5 Preview
Task & likeness8.5/104.0/10
Anatomy & construction6.5/104.5/10
Space & placement7.0/103.0/10
Light & reflections5.5/103.5/10
Purposeful detail6.5/104.0/10
Artistry & character6.5/103.5/10
Materials & effects6.5/103.0/10
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

Qwen3.8-27B

The pelican’s oversized bill, straw hat and flowing scarf create a distinctive focal character against a well-composed coastal sunset. Wheels and frame are readable, but the wings do not reach the steering, overlapping legs obscure separate pedal contacts, and the crank linkage is unclear. Clean vector shapes and restrained shading support an appealing illustration with limited physical depth.

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

CategoryQwen3.8-27BStep 5 Preview
Task & likeness8.0/103.0/10
Anatomy & construction5.5/103.0/10
Space & placement7.0/103.0/10
Light & reflections7.0/103.0/10
Purposeful detail7.5/104.0/10
Artistry & character7.5/103.0/10
Materials & effects6.5/103.0/10
Visual integrity8.0/102.0/10

Benchmark task

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