STEALTHMARK · HEAD TO HEAD

All comparisons ↗

Inkling-Small vs Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) leads Inkling-Small by 35.0 points on StealthMark, 62.6 to 27.6. It leads on all 9 benchmarks. Union Alpha scores higher on Newton's Mirror, with its largest category advantage in reflections.

See the difference ↘
All results ↗
Model guide ↗

Overall StealthMark

27.6/100Overall rank 43/53
62.6/100Overall rank 8/53

General Intelligence

25.5/100
58.7/100

Visual Intelligence

31.1/100
69.6/100

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
  • Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)Not measured

Open or local

Open modelApache 2.0

BENCHMARK BY BENCHMARK

Where each model leads

  • Inkling-Small
  • Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)
  • Other models
  • Median

DATASHEET

Specs and pricing

Inkling-Small takes 4× more context. Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) scores higher on StealthMark.

  • Inkling-Small
  • Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)
  • Other models
  • Median
Context windowTokens per request, max output below
Inkling-Small: 1M4× largerPareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM): 262K131K max output
Output speedTokens per second, our measurement
Inkling-Small: Not measuredPareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM): 26tokens/s
SizeTotal and active parameters
Inkling-Small: 276B12B activePareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM): Not disclosed
Developer
Thinking Machines LabCircuit & Chisel / Unbiased
Released
Not published20 Sep 2026
Origin
Not publishedUnited States
Input → output
Text, image, audio → textText, image → text
Architecture
42-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 audioNot published

Full datasheets, every price and sourceInkling-Small ↗Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) ↗

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.

Inkling-Small 30.2/100Rank 29/49

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

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) 78.7/100Rank 1/49

Survived 2 of 3
Foresight: complete playing field
Level 319 ptsTurn 20 / 20

Points per level

LevelInkling-SmallPareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)
Level 1Second Thoughts
9 ptsOut of lives
27 ptsSurvived
Level 2Pressure Pot
6 ptsOut of lives
12 ptsOut of lives
Level 3Hide or Flight
8 ptsOut of lives
19 ptsSurvived

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.

Inkling-Small 34.7/100Rank 34/50

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

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) 36.5/100Rank 32/50

Cleared 4 of 5
Latch: complete playing field
Level 519.4 ptsTurn 33 / 40

Points per level

LevelInkling-SmallPareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
9.8 ptsCleared
9.8 ptsCleared
Level 3Relay vault
2.8 ptsFinished early
14.2 ptsClearedFinished early · 4 beacons left
Level 4Passing bay
8 ptsFinished early
9.2 ptsFinished early
Level 5Second thoughts
18 ptsCleared
19.4 ptsCleared

Newton's Mirror

Same task instructions

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

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) 52.0/100 Rank 7/53

Union Alpha: Newton's Mirror SVG benchmark

The judgment

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.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)

Recognizable near-field sphere placement and bilateral convergence provide limited optical structure. Broad gray regions interrupt lateral sightlines and truncate side-floor images. Missing outer chains, absent nested sphere imagery and incomplete planar and curved emitter mappings dominate the reflection system.

Score breakdown

CategoryInkling-SmallPareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)
State at 600 ms14.0/2524.0/25
Camera and placement3.0/1513.5/15
Reflections8.0/12034.5/120
Materials and scale2.5/1010.0/10
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.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) 76.5/100 Rank 7/53

Union Alpha: Selfie SVG benchmark

The judgment

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.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)

A strong futuristic emblem with clear hierarchy, controlled negative space and distinctive typography. Silver and gunmetal layers give the sweeping ribbons convincing graphic depth, while red accents remain restrained. Ambiguous concealed connections and occasionally abrupt highlight bands limit the construction and material realism.

Score breakdown

CategoryInkling-SmallPareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)
Task & likeness2.5/109.0/10
Anatomy & construction1.5/107.5/10
Space & placement3.5/108.5/10
Light & reflections3.0/107.5/10
Purposeful detail4.5/107.0/10
Artistry & character3.0/107.5/10
Materials & effects3.5/107.5/10
Visual integrity2.0/107.0/10

Benchmark task

Inkling-Small

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

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)

An interwoven metallic emblem tests ribbon geometry, overlapping forms and the appearance of reflective metal against a dark background.

Mona Lisa

Different task instructions

Precise visual recall and portrait reconstruction.

The judgment

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.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)

The centered portrait, folded hands, dark clothing and hazy landscape clearly evoke the Mona Lisa. Muted colors and soft facial shading create a coherent illustration, but replication remains schematic: the face is broad, eyes outlined, smile simplified and overlapping fingers poorly resolved. Angular landscape shapes and flat fabric limit depth and painterly richness.

Score breakdown

CategoryInkling-SmallPareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)
Task & likeness2.5/106.5/10
Anatomy & construction2.5/105.0/10
Space & placement3.0/106.5/10
Light & reflections2.5/106.0/10
Purposeful detail2.0/105.5/10
Artistry & character2.5/106.5/10
Materials & effects2.0/105.0/10
Visual integrity3.0/107.0/10

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

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.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)

A cohesive sunset palette, sweeping shoreline and playful animals make an appealing beach illustration. The sun’s reflection establishes a clear focal axis, balanced by the palm and umbrella. Clean silhouettes and small decorative features reward inspection, though simplified anatomy, repetitive reflection bars and generic oval shadows limit volume and naturalism.

Score breakdown

CategoryInkling-SmallPareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)
Task & likeness3.5/109.0/10
Anatomy & construction3.0/107.0/10
Space & placement3.0/108.0/10
Light & reflections3.0/107.0/10
Purposeful detail3.0/107.0/10
Artistry & character3.0/106.0/10
Materials & effects2.5/106.5/10
Visual integrity2.0/107.0/10

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.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM) 63.3/100 Rank 8/53

Union Alpha: Ground Truth SVG benchmark

The judgment

Inkling-Small

This benchmark has not been assessed yet.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)

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

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.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)

Expressive animals, a readable riding arrangement and sweeping spray create an appealing cartoon. The raccoon sits convincingly on the dog's back, but the dog's paw does not clearly grip the handlebar, whose connection to the hull is unresolved. Flat lighting provides little volume or reflection. A dark rectangular strip interrupts the upper-right background.

Score breakdown

CategoryInkling-SmallPareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)
Task & likeness4.0/108.0/10
Anatomy & construction3.0/106.5/10
Space & placement4.0/107.5/10
Light & reflections5.0/105.5/10
Purposeful detail5.5/107.0/10
Artistry & character4.0/107.0/10
Materials & effects5.0/107.5/10
Visual integrity5.0/107.0/10

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

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.

Pareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)

The pelican is immediately recognizable, with an expressive face, convincing bill silhouette and appealing pastel composition. Both legs connect to feet, and the wing reaches the handlebar. Bicycle construction is less resolved: the far pedal lacks a clear crank connection and no chain is visible. Clean contours and orderly spokes support the illustration, while flat lighting and simplified surfaces limit depth.

Score breakdown

CategoryInkling-SmallPareto 26.9 (Union of GPT-6 Astra + DeepSeek + GLM)
Task & likeness4.0/108.0/10
Anatomy & construction3.0/106.0/10
Space & placement3.0/107.0/10
Light & reflections3.0/106.0/10
Purposeful detail4.5/106.0/10
Artistry & character5.0/106.0/10
Materials & effects4.5/106.0/10
Visual integrity3.0/108.5/10

Benchmark task

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