STEALTHMARK · HEAD TO HEAD

All comparisons ↗

GPT-6.1 Terra vs Hy4 preview

GPT-6.1 Terra leads Hy4 preview by 16.5 points on StealthMark, 60.4 to 43.9. It leads on all 9 benchmarks. GPT-6.1 Terra scores higher on Ground Truth.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

60.4/100Overall rank 9/54
43.9/100Overall rank 31/54

General Intelligence

58.0/100
41.3/100

Visual Intelligence

64.9/100
48.6/100

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

  • Hy4 preview

Open or local

Open modelApache-2.0

BENCHMARK BY BENCHMARK

Where each model leads

  • GPT-6.1 Terra
  • Hy4 preview
  • Other models
  • Median

DATASHEET

Specs and pricing

GPT-6.1 Terra scores higher on StealthMark. Hy4 preview is 2× cheaper per output token. Both take about 1.05M tokens of context.

  • GPT-6.1 Terra
  • Hy4 preview
  • Other models
  • Median
PricePer 1M output tokens, input below
GPT-6.1 Terra: $5$1 inputHy4 preview: $2.502× cheaper$0.83 input
Context windowTokens per request, max output below
GPT-6.1 Terra: 1.05MTied128K max outputHy4 preview: 1MTied64K max output
Output speedTokens per second, our measurement
GPT-6.1 Terra: Not measuredHy4 preview: 64tokens/s
SizeTotal and active parameters
GPT-6.1 Terra: Not disclosedHy4 preview: 770B49B active
Developer
OpenAITencent Hy Team
Released
26 Oct 202628 Aug 2026
Origin
United StatesChina
Input → output
Text, image → textText → text
Architecture
Not publishedMoE; 78-layer backbone with Gated DSA, 256 routed experts + 1 shared expert, top-8 routed experts active, plus native MTP layer

Full datasheets, every price and sourceGPT-6.1 Terra ↗Hy4 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.

GPT-6.1 Terra 62.0/100Rank 9/50

Survived 2 of 3
Foresight: complete playing field
Level 315 ptsTurn 20 / 20

Hy4 preview 53.1/100Rank 15/50

Survived 1 of 3
Foresight: complete playing field
Level 311 ptsTurn 16 / 20

Points per level

LevelGPT-6.1 TerraHy4 preview
Level 1Second Thoughts
21 ptsSurvived
26 ptsSurvived
Level 2Pressure Pot
9 ptsOut of lives
6 ptsOut of lives
Level 3Hide or Flight
15 ptsSurvived
11 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.

GPT-6.1 Terra 77.7/100Rank 12/51

Cleared 5 of 5
Latch: complete playing field
Level 520.6 ptsTurn 27 / 40

Hy4 preview 35.6/100Rank 34/51

Cleared 3 of 5
Latch: complete playing field
Level 518.2 ptsTurn 19 / 40

Points per level

LevelGPT-6.1 TerraHy4 preview
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
9.8 ptsCleared
6 ptsFinished early
Level 3Relay vault
19.2 ptsCleared
18 ptsCleared
Level 4Passing bay
17.8 ptsClearedFinished early · 3 beacons left
15.8 ptsClearedFinished early · 3 beacons left
Level 5Second thoughts
20.6 ptsCleared
18.2 ptsFinished early

Newton's Mirror

Same task instructions

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

The judgment

GPT-6.1 Terra

The sphere, airborne black and first reflected silhouettes preserve important geometry within the portrait frame. Paint deformation, contact order and forward transport are supported, with a small crimson endpoint overshoot. Major optical omissions dominate: short finite reflection sequences and smooth bands replace extensive, spatially ordered mirror imagery. Dark gloss and crisp upper emitter shapes provide partial material credit.

Hy4 preview

The central sphere, compact airborne black bead and broad paint ordering are supported. Contact and transport history remain partly unresolved. The optical system provides only isolated first images and light patches, with little evidence of the extensive ordered reflection structure. Rounded metal and paint gloss are more successful than the reflected contents.

Score breakdown

CategoryGPT-6.1 TerraHy4 preview
State at 600 ms23.0/2518.0/25
Camera and placement11.0/1511.5/15
Reflections33.0/12029.5/120
Materials and scale7.5/107.5/10
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

GPT-6.1 Terra

A balanced serpent emblem with six attached outward heads, delicate scales and softly rounded bodies. Its compact silhouette reads clearly, but central routing does not establish every alternating crossing or crisp inward corner. Repetitive texture, limited facial anatomy and weak contact shadows constrain its organic character.

Hy4 preview

A cohesive, appealing illustration with a serene face, luminous palette and complete framing. The separated head and torso weaken anatomical coherence, while crowded wing attachments and angular sash folds limit the requested suppleness. Clean execution and decorative detail exceed the clarity of the underlying construction.

Score breakdown

CategoryGPT-6.1 TerraHy4 preview
Task & likeness7.5/106.0/10
Anatomy & construction6.5/105.5/10
Space & placement6.5/107.5/10
Light & reflections6.0/107.0/10
Purposeful detail6.5/106.5/10
Artistry & character6.5/105.0/10
Materials & effects6.0/107.0/10
Visual integrity6.0/105.5/10

Benchmark task

GPT-6.1 Terra

A pale serpent woven into an interlocking knot tests connected anatomy, repeated structure and readable overlaps. Organic surface detail must work with the overall geometry.

Hy4 preview

A young blue dragon floating among clouds tests curved anatomy, expressive features and the contrast between soft clouds and glossy surfaces.

Mona Lisa

Different task instructions

Precise visual recall and portrait reconstruction.

The judgment

GPT-6.1 Terra

A recognizable interpretation with developed facial shading, shaped sleeve folds and purposeful small features. These strengths support modest upward calibration, while the frontal torso, elongated proportions, schematic finger joints and unresolved seated support keep replication and construction uneven.

Hy4 preview

A recognizable but basic Mona Lisa interpretation with a cohesive earthy palette and atmospheric hills. The rigid neck, generic face and unresolved folded-hand pose limit replication and anatomy. Large silhouettes and layer boundaries are comparatively clear, warranting stronger integrity credit than the sparse detail and simplified construction.

Score breakdown

CategoryGPT-6.1 TerraHy4 preview
Task & likeness5.5/102.5/10
Anatomy & construction6.5/102.5/10
Space & placement4.5/102.5/10
Light & reflections6.5/101.5/10
Purposeful detail6.0/102.5/10
Artistry & character6.0/101.5/10
Materials & effects6.0/102.5/10
Visual integrity6.5/104.0/10

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

GPT-6.1 Terra

A coherent sunset illustration with clear prompt fulfillment, controlled framing and varied, purposeful small details. Simplified joints, frontal staging, flattened animal forms and weakly directional shadows limit its anatomical, spatial and physical sophistication.

Hy4 preview

A cheerful beach illustration fulfills the visible subject requirements with coherent foreground construction and a harmonious sunset palette. Shallow depth, schematic reflections, ambiguous blue airborne forms and exposed geometry beneath the sunset limit its finish. Small accessories are readable, but materials and surface detail remain simplified.

Score breakdown

CategoryGPT-6.1 TerraHy4 preview
Task & likeness8.5/108.5/10
Anatomy & construction7.5/105.5/10
Space & placement6.5/106.5/10
Light & reflections5.5/105.0/10
Purposeful detail7.5/106.0/10
Artistry & character7.0/105.5/10
Materials & effects6.5/105.0/10
Visual integrity7.5/106.0/10

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

GPT-6.1 Terra

This benchmark has not been assessed yet.

Hy4 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

GPT-6.1 Terra

A coherent cartoon with expressive animals, readable steering and a clear wake. Fine facial and hull details reward inspection. The main limits are the boatlike craft, simplified joints, weakly braced frontal raccoon and minimal directional lighting or material depth.

Hy4 preview

A lively cartoon communicates the animal pair traveling across water. Curving waves, rising spray and locally modeled gray forms strengthen motion and volume. Ambiguous seating, simplified joints and unclear control routing limit construction. Competing focal elements, decorative foam and inconsistent illumination leave the finish uneven.

Score breakdown

CategoryGPT-6.1 TerraHy4 preview
Task & likeness6.5/107.0/10
Anatomy & construction5.5/104.0/10
Space & placement6.0/106.0/10
Light & reflections5.0/106.5/10
Purposeful detail7.0/105.5/10
Artistry & character6.0/104.5/10
Materials & effects6.0/105.5/10
Visual integrity7.0/105.0/10

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

Same task instructions

Subject construction and spatial relationships.

The judgment

GPT-6.1 Terra

A coherent pelican cyclist with expressive proportions, purposeful feather detail and developed bicycle machinery. Continuous body contours, aligned wheels and readable pedal contacts support the scene. Simplified articulation, shallow modelling and an unresolved steering connection limit construction and placement.

Hy4 preview

A recognizable, playful pelican cyclist with readable wheels, accessories and landscape layers. Form shading and ground shadows provide competent depth. Angular legs, uncertain pedal engagement and absent steering contact limit the riding construction; crowded overlaps and simple surfaces constrain refinement.

Score breakdown

CategoryGPT-6.1 TerraHy4 preview
Task & likeness7.0/107.5/10
Anatomy & construction7.0/104.5/10
Space & placement6.0/106.5/10
Light & reflections5.5/107.0/10
Purposeful detail7.0/106.0/10
Artistry & character6.5/105.0/10
Materials & effects5.5/105.0/10
Visual integrity6.0/107.0/10

Benchmark task

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