STEALTHMARK · HEAD TO HEAD

All comparisons ↗

Claude Fable 5.2 vs Kolibri-1

Claude Fable 5.2 leads Kolibri-1 by 66.2 points on StealthMark, 79.8 to 13.6. It leads on all 7 benchmarks with results for both. Claude Fable 5.2 scores higher on Newton's Mirror, with its largest category advantage in reflections.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

79.8/100Overall rank 1/58
13.6/100Overall rank 57/58

General Intelligence

81.7/100
12.3/100

Visual Intelligence

76.0/100
15.0/100

Reasoning efficiency

How many reasoning tokens a model needs for the quality it delivers.

Single-step reasoning How much it reasons to understand a task

Multi-step reasoning How much it keeps reasoning as a task goes on

  • Kolibri-1Multi-step reasoning not measured

Open or local

Local modelOpen model · Apache-2.0

BENCHMARK BY BENCHMARK

Where each model leads

  • Claude Fable 5.2
  • Kolibri-1
  • Other models
  • Median

DATASHEET

Specs and pricing

  • Claude Fable 5.2
  • Kolibri-1
  • Other models
  • Median
Context windowTokens per request, max output below
Claude Fable 5.2: Not publishedKolibri-1: 1M
Output speedTokens per second, our measurement
Claude Fable 5.2: Not measuredKolibri-1: 166tokens/s
SizeTotal and active parameters
Claude Fable 5.2: Not disclosedKolibri-1: 78.1B3.46B active
Developer
Not publishedAleph Alpha
Released
Not published3 Oct 2026
Origin
Not publishedGermany
Input → output
Not publishedText → text
Architecture
Not publishedMixture-of-Experts with 4:1 sliding-window-attention to global-attention pattern

Full datasheets, every price and sourceKolibri-1 ↗

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.

Claude Fable 5.2 77.2/100Rank 2/54

Survived 3 of 3
Foresight: complete playing field
Level 318 ptsTurn 20 / 20

Points per level

LevelClaude Fable 5.2Kolibri-1
Level 1Second Thoughts
27 ptsSurvived
No run
Level 2Pressure Pot
12 ptsSurvived
No run
Level 3Hide or Flight
18 ptsSurvived
No run

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.

Claude Fable 5.2 100.0/100Rank 1/55

Cleared 5 of 5
Latch: complete playing field
Level 520.8 ptsTurn 26 / 40

Points per level

LevelClaude Fable 5.2Kolibri-1
Level 1Hold the door
5.2 ptsCleared
No run
Level 2Crossfeed
10.2 ptsClearedFinished early · 1 beacon left
No run
Level 3Relay vault
19.2 ptsCleared
No run
Level 4Passing bay
18.2 ptsClearedFinished early · 3 beacons left
No run
Level 5Second thoughts
20.8 ptsCleared
No run

Newton's Mirror

Same task instructions

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

The judgment

Claude Fable 5.2

The composition closely preserves the principal camera relationships. Visible cyan contact, elongated crimson coating and separate airborne black paint support the required interaction sequence, with a small crimson transport overshoot. Correct broad mirror placement coexists with major omissions in nested spherical imagery and simplified emitter mapping. Core score: 105.5/170; floor-damage reward: 0/10; combined total: 105.5/180.

Kolibri-1

A segmented gray perimeter surrounds a black field containing one large, low-positioned gray sphere and two small colored circles. Rounded volume and restrained paint scale earn partial material credit. The image does not demonstrate the required optical structure, while paint history, airborne black and settled support remain largely unresolved.

Score breakdown

CategoryClaude Fable 5.2Kolibri-1
State at 600 ms24.0/259.5/25
Camera and placement15.0/151.5/15
Reflections80.0/1200.0/120
Materials and scale8.5/103.0/10
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

Claude Fable 5.2

The compact mantle, eight curling rays and quiet face closely realize the requested living emblem. Controlled arm contours, coherent overlaps and restrained ornament give the design character. Shallow depth, ribbon-like arm modeling and repeated highlight bands limit anatomical and lighting complexity without compromising the intact silhouette.

Kolibri-1

A restrained ivory, teal and green palette gives the image simple graphic coherence. Execution remains rudimentary: detached oval wings, ambiguous central anatomy and unclear feeding contact prevent a convincing hummingbird scene. The requested feather structure, botanical detail, lighting and enclosing arch are largely unresolved.

Score breakdown

CategoryClaude Fable 5.2Kolibri-1
Task & likeness9.0/102.5/10
Anatomy & construction7.5/102.0/10
Space & placement7.0/103.0/10
Light & reflections7.0/102.0/10
Purposeful detail6.5/101.5/10
Artistry & character6.5/101.5/10
Materials & effects7.0/102.0/10
Visual integrity8.5/105.5/10

Benchmark task

Claude Fable 5.2

A coral squid with spreading arms tests connected anatomy, balanced composition and soft surface shading.

Kolibri-1

A jewel-coloured hummingbird hovering in flight tests avian anatomy, delicate structure and iridescent surface detail.

Mona Lisa

Different task instructions

Precise visual recall and portrait reconstruction.

The judgment

Claude Fable 5.2

A developed illustrative portrait with controlled facial shading, connected limbs, layered scenery and distinct aged-paint and gold effects. Its frontal torso, conspicuous smile and flattened fingers limit replication and anatomical precision. Decorative finish is stronger than the underlying seated construction.

Kolibri-1

The candidate offers a clean, highly simplified frontal portrait with warm skin and dark hair. Its oversized oval face, schematic features and unresolved body provide little Mona Lisa likeness. Missing crossed hands, clothing detail and landscape remove essential parts of the original composition; smooth gradients supply only limited volume.

Score breakdown

CategoryClaude Fable 5.2Kolibri-1
Task & likeness6.5/100.5/10
Anatomy & construction6.0/100.5/10
Space & placement6.0/100.0/10
Light & reflections6.5/100.0/10
Purposeful detail6.0/100.0/10
Artistry & character6.5/100.5/10
Materials & effects7.5/100.5/10
Visual integrity7.5/100.5/10

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

Claude Fable 5.2

A lively sunset beach with clear playful activity, purposeful small details and developed water reflections. Layering and supported placements hold the scene together. Schematic anatomy and the turtle's overhead viewpoint limit construction, space and artistry, while atmospheric and surface effects show greater refinement.

Kolibri-1

A recognizable sunset beach has an appealing purple-orange gradient and clean geometric edges. Birds and animals remain unresolved, terrain obscures the relationship between shore and water, and reflections are absent. Sparse detail, flat materials and limited activity keep the drawing elementary despite its tidy rendering.

Score breakdown

CategoryClaude Fable 5.2Kolibri-1
Task & likeness9.0/102.5/10
Anatomy & construction8.0/100.5/10
Space & placement7.5/103.0/10
Light & reflections8.5/101.0/10
Purposeful detail8.5/100.5/10
Artistry & character8.0/101.0/10
Materials & effects7.5/100.5/10
Visual integrity8.0/103.0/10

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

Claude Fable 5.2

This benchmark has not been assessed yet.

Kolibri-1

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

Claude Fable 5.2

A controlled action illustration with expressive animals, demonstrated riding and steering, and an energetic layered wake. Purposeful small features and articulated poses strengthen the drawing. Simplified extremities, shallow character volume and limited physical lighting remain its principal constraints.

Kolibri-1

Clean shapes and a readable coral platform provide basic graphic clarity, but the animal construction remains fragmented. A recognizable raccoon riding the dog's back, connected steering and convincing water contact are not demonstrated. Sparse detail, flat subjects and absent movement effects support retaining the baseline category assessments.

Score breakdown

CategoryClaude Fable 5.2Kolibri-1
Task & likeness8.5/101.5/10
Anatomy & construction8.0/101.0/10
Space & placement7.5/101.0/10
Light & reflections7.0/100.0/10
Purposeful detail8.5/100.5/10
Artistry & character8.0/100.5/10
Materials & effects7.5/100.5/10
Visual integrity8.0/103.5/10

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

Claude Fable 5.2

A strong illustration with an expressive pelican, readable riding contacts and carefully constructed bicycle details. Layered plumage, basket weave and coastal scenery reward inspection. Continuous attachments support the action, while limited visible leg articulation, modest volume and weakly directional foreground lighting constrain physical sophistication.

Kolibri-1

Clean shapes and aligned wheels establish a readable schematic scene. Pelican likeness, riding support and bicycle mechanics remain substantially unresolved. Sparse detail and flat surfaces limit expression and volume. Strong edge and fill integrity is a separate success and does not repair the construction.

Score breakdown

CategoryClaude Fable 5.2Kolibri-1
Task & likeness9.0/101.5/10
Anatomy & construction8.0/101.5/10
Space & placement7.5/101.5/10
Light & reflections6.5/103.0/10
Purposeful detail8.0/101.5/10
Artistry & character7.5/101.0/10
Materials & effects7.0/101.0/10
Visual integrity8.0/103.5/10

Benchmark task

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