STEALTHMARK · HEAD TO HEAD

All comparisons ↗

Glyph Cluster vs Inkling-Small

Inkling-Small leads Glyph Cluster by 8.7 points on StealthMark, 27.6 to 18.9. It leads on 6 of 9 benchmarks, including both games, and Glyph Cluster on 3. Inkling-Small scores higher on Newton's Mirror, with its largest category advantage in state at 600 ms.

See the difference ↘
Model guide ↗
All results ↗

Overall StealthMark

18.9/100Overall rank 48/53
27.6/100Overall rank 43/53

General Intelligence

12.0/100
25.5/100

Visual Intelligence

29.8/100
31.1/100

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

  • Inkling-Small

Open or local

Open modelApache 2.0

BENCHMARK BY BENCHMARK

Where each model leads

  • Glyph Cluster
  • Inkling-Small
  • Other models
  • Median

DATASHEET

Specs and pricing

  • Glyph Cluster
  • Inkling-Small
  • Other models
  • Median
Context windowTokens per request, max output below
Glyph Cluster: Not publishedInkling-Small: 1M
Output speedTokens per second, our measurement
Glyph Cluster: 69tokens/sInkling-Small: Not measured
SizeTotal and active parameters
Glyph Cluster: Not disclosedInkling-Small: 276B12B active
Developer
Not publishedThinking Machines Lab
Input → output
Not publishedText, image, audio → text
Architecture
Not published42-layer decoder-only transformer with sparse mixture-of-experts feed-forward (6 of 256 routed experts plus 2 shared per token); hybrid local and global attention; natively multimodal with a hierarchical patch image encoder and discrete-token audio

Full datasheets, every price and sourceInkling-Small ↗

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.

Glyph Cluster 6.4/100Rank 46/49

Survived 0 of 3
Foresight: complete playing field
Level 33 ptsTurn 5 / 20

Inkling-Small 30.2/100Rank 29/49

Survived 0 of 3
Foresight: complete playing field
Level 38 ptsTurn 17 / 20

Points per level

LevelGlyph ClusterInkling-Small
Level 1Second Thoughts
3 ptsOut of lives
9 ptsOut of lives
Level 2Pressure Pot
3 ptsOut of lives
6 ptsOut of lives
Level 3Hide or Flight
3 ptsOut of lives
8 ptsOut of lives

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.

Glyph Cluster 3.6/100Rank 46/50

Cleared 1 of 5
Latch: complete playing field
Level 59 ptsTurn 25 / 40

Inkling-Small 34.7/100Rank 34/50

Cleared 3 of 5
Latch: complete playing field
Level 518 ptsTurn 40 / 40

Points per level

LevelGlyph ClusterInkling-Small
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
2 ptsOut of turns
9.8 ptsCleared
Level 3Relay vault
4.2 ptsFinished early
2.8 ptsFinished early
Level 4Passing bay
1 ptsOut of turns
8 ptsFinished early
Level 5Second thoughts
9 ptsFinished early
18 ptsCleared

Newton's Mirror

Same task instructions

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

The judgment

Glyph Cluster

Rounded sphere volume, compact colored forms and crimson-below-cyan ordering provide limited agreement. The visible paint does not establish the required interaction history, the sphere and black blob are displaced from the prescribed projection, and most of the defining mirror-room structure is missing or unresolved. The faint reflection echoes earn limited optical credit.

Inkling-Small

The sphere's apparent floor support, separate dark feature and broad paint ordering justify partial state credit. The major weakness is the largely missing mirror system: isolated side images and faint floor echoes do not establish repeated virtual rooms or curved spherical mapping. Downward framing, the oversized frontal panel, unresolved paint transport and satin surface appearance remain distinct limitations.

Score breakdown

CategoryGlyph ClusterInkling-Small
State at 600 ms5.0/2514.0/25
Camera and placement3.0/153.0/15
Reflections8.0/1208.0/120
Materials and scale3.0/102.5/10
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

Glyph Cluster

A charming palette and three expressive companions make the concept recognizable. The dominant paper enclosure overwhelms the small group, the plum character lacks a clear brace silhouette, and the cuddle remains difficult to trace. Sparse furnishings, ambiguous folds and generalized shading limit the promised lived-in richness and tactile depth.

Inkling-Small

The faces and dripping caps communicate a cute mushroom scene, but simple construction and weak spatial relationships limit the result. The defining foot-stamping action is unresolved, the flattering portrait is smaller than its painter, and the requested material contrasts are only partly conveyed.

Score breakdown

CategoryGlyph ClusterInkling-Small
Task & likeness3.5/102.5/10
Anatomy & construction2.5/101.5/10
Space & placement4.0/103.5/10
Light & reflections3.5/103.0/10
Purposeful detail4.5/104.5/10
Artistry & character3.0/103.0/10
Materials & effects2.0/103.5/10
Visual integrity3.0/102.0/10

Benchmark task

Glyph Cluster

Punctuation-shaped companions gather inside a curled paper home. The scene tests expressive silhouettes, close interaction and the coherence of a detailed miniature environment.

Inkling-Small

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

Mona Lisa

Different task instructions

Precise visual recall and portrait reconstruction.

The judgment

Glyph Cluster

The portrait has a continuous body silhouette, attached sleeves and hands, and simple landscape depth. Its oversized pale head enclosure nearly erases the expression, while separated fists, ochre clothing and frontal construction weaken Mona Lisa likeness. Sparse detail, flat materials and ambiguous facial layers leave the drawing seriously unresolved.

Inkling-Small

A recognizable but rudimentary Mona Lisa interpretation. Facial shading and subdued colors provide coherence, while the frontal blocklike torso, unresolved wrists, fingerless hands and shallow landscape substantially limit likeness, construction and finish.

Score breakdown

CategoryGlyph ClusterInkling-Small
Task & likeness2.0/102.5/10
Anatomy & construction0.5/102.5/10
Space & placement2.0/103.0/10
Light & reflections1.0/102.5/10
Purposeful detail1.0/102.0/10
Artistry & character0.5/102.5/10
Materials & effects1.0/102.0/10
Visual integrity3.0/103.0/10

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

Glyph Cluster

A cheerful, recognizable beach illustration with playful accessories and mostly clean silhouettes. Sunset colors and soft glow supply atmosphere, but abbreviated anatomy, unexplained airborne animals, weak ground contact and absent water reflections limit physical coherence. Small details add interest without consistently resolving construction or materials.

Inkling-Small

A recognizable sunset beach with a clear focal sun and appealing color transitions. Connected fish silhouettes and layered scenery establish basic coherence. Primitive figures, ambiguous recreation, sparse descriptive detail and weak grounding limit construction and depth. The aligned reflection supplies a useful lighting cue, but shadows and material effects remain underdeveloped.

Score breakdown

CategoryGlyph ClusterInkling-Small
Task & likeness6.0/103.5/10
Anatomy & construction4.0/103.0/10
Space & placement5.0/103.0/10
Light & reflections4.0/103.0/10
Purposeful detail5.5/103.0/10
Artistry & character4.0/103.0/10
Materials & effects4.0/102.5/10
Visual integrity5.0/102.0/10

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

Glyph Cluster

This benchmark has not been assessed yet.

Inkling-Small

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

Glyph Cluster

Recognizable animals and clean shapes give the scene charm, but the requested steering action is visibly incomplete. Simplified joints, weak vehicle construction and uncertain water contact limit physical coherence. Flat lighting, sparse functional detail and a small focal group keep the illustration basic despite its tidy edges.

Inkling-Small

Recognizable animals, readable accessories and an appealing sunset support a coherent basic cartoon. Rounded construction leaves joints, riding support and steering unresolved. Decorative gradients contribute atmosphere, but directional lighting, distinct materials and convincing water displacement are weakly demonstrated.

Score breakdown

CategoryGlyph ClusterInkling-Small
Task & likeness5.0/104.0/10
Anatomy & construction3.0/103.0/10
Space & placement4.0/104.0/10
Light & reflections2.5/105.0/10
Purposeful detail3.5/105.5/10
Artistry & character3.0/104.0/10
Materials & effects3.0/105.0/10
Visual integrity5.5/105.0/10

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

Glyph Cluster

A readable cartoon with open framing, aligned bicycle wheels and modest atmospheric effects. Pelican likeness is approximate, while the narrow neck, crowded feet, absent steering contact and unclear drivetrain weaken the riding action. Clean shapes help presentation, but shallow space and inconsistent shading limit the finish.

Inkling-Small

The pelican is immediately recognizable, and its sweeping neck, shaped bill and orderly spokes provide character. Riding remains only partially realized: legs, pedal engagement, steering and seat support cannot be read confidently. Controlled graphic effects add atmosphere, but simplified construction and weak grounding keep the result basic.

Score breakdown

CategoryGlyph ClusterInkling-Small
Task & likeness6.0/104.0/10
Anatomy & construction3.0/103.0/10
Space & placement6.0/103.0/10
Light & reflections3.5/103.0/10
Purposeful detail4.0/104.5/10
Artistry & character3.0/105.0/10
Materials & effects4.5/104.5/10
Visual integrity6.0/103.0/10

Benchmark task

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