STEALTHMARK · HEAD TO HEAD

All comparisons ↗

GLM-5.3 vs GPT-5.6 Terra

GLM-5.3 leads GPT-5.6 Terra by 7.1 points on StealthMark, 52.9 to 45.8. It leads on 7 of 9 benchmarks, including Foresight, and GPT-5.6 Terra on 2, including Latch. GLM-5.3 scores higher on Newton's Mirror, with its largest category advantage in reflections.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

52.9/100Overall rank 16/58
45.8/100Overall rank 30/58

General Intelligence

46.5/100
41.9/100

Visual Intelligence

64.1/100
52.7/100

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

  • GPT-5.6 Terra

Cost efficiency

How much Performance you get for your money.

Open or local

Open modelGLM-5.3 License

BENCHMARK BY BENCHMARK

Where each model leads

  • GLM-5.3
  • GPT-5.6 Terra
  • Other models
  • Median

DATASHEET

Specs and pricing

GLM-5.3 scores higher on StealthMark and is 2.7× cheaper per output token. GPT-5.6 Terra runs 18% faster in our speed test. Both take about 1M tokens of context.

  • GLM-5.3
  • GPT-5.6 Terra
  • Other models
  • Median
PricePer 1M output tokens, input below
GLM-5.3: $4.402.7× cheaper$1.40 inputGPT-5.6 Terra: $12$2 input
Context windowTokens per request, max output below
GLM-5.3: 1MTied128K max outputGPT-5.6 Terra: 1.05MTied128K max output
Output speedTokens per second, our measurement
GLM-5.3: 86tokens/sGPT-5.6 Terra: 10218% fastertokens/s
SizeTotal and active parameters
GLM-5.3: 753BAll activeGPT-5.6 Terra: Not disclosed
Developer
Z.AIOpenAI
Released
17 Aug 20269 Jul 2026
Origin
ChinaUnited States
Input → output
Text → textText, image → text
Architecture
GLM MoE with DSA sparse attentionNot published

Full datasheets, every price and sourceGLM-5.3 ↗GPT-5.6 Terra ↗

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.

GLM-5.3 43.6/100Rank 25/54

Survived 1 of 3
Foresight: complete playing field
Level 34 ptsTurn 7 / 20

GPT-5.6 Terra 30.1/100Rank 36/54

Survived 0 of 3
Foresight: complete playing field
Level 36 ptsTurn 7 / 20

Points per level

LevelGLM-5.3GPT-5.6 Terra
Level 1Second Thoughts
27 ptsSurvived
5 ptsOut of lives
Level 2Pressure Pot
6 ptsOut of lives
9 ptsOut of lives
Level 3Hide or Flight
4 ptsOut of lives
6 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.

GLM-5.3 25.6/100Rank 44/55

Cleared 2 of 5
Latch: complete playing field
Level 512.2 ptsTurn 14 / 40

GPT-5.6 Terra 80.0/100Rank 11/55

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

Points per level

LevelGLM-5.3GPT-5.6 Terra
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
9 ptsCleared
9.6 ptsCleared
Level 3Relay vault
7.6 ptsFinished early
19.2 ptsCleared
Level 4Passing bay
3.2 ptsFinished early
18.2 ptsClearedFinished early · 3 beacons left
Level 5Second thoughts
12.2 ptsFinished early
20.6 ptsCleared

Newton's Mirror

Same task instructions

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

The judgment

GLM-5.3

The central sphere, isolated airborne black blob and crimson leading-end placement are well established. Recent cyan contact remains only partially demonstrated. The dominant weakness is optical: sparse shaded sphere rows, missing outer chains and diffuse highlights replace extensive mirror repetition and curved nested imagery.

GPT-5.6 Terra

The real sphere and airborne black blob are substantially too low in the complete frame after accounting for aspect ratio. The first wall and floor reflections have incompatible placement, relative size or silhouette. Both lateral regions lack the references' receding lane-and-chain organization. The few recognizable mirror images provide limited optical structure. The main, side and floor spheres replace nested mirror imagery with smooth bands, while sparse planar light patches and broad spherical highlights fail to establish coherent repeated emitter mapping.

Score breakdown

CategoryGLM-5.3GPT-5.6 Terra
State at 600 ms22.5/2524.0/25
Camera and placement13.0/156.0/15
Reflections37.0/1207.0/120
Materials and scale7.0/105.5/10
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

GLM-5.3

A luminous organic creature meets a geometric reflection in a carefully framed oval mirror. Shared glowing cores and fingertip contact make the introspective symbolism clear. Gold, cream and blue establish a controlled hierarchy. The creature’s lower anatomy and shoulder transition remain loosely defined, while ground contact and reflected depth are mostly symbolic. Elegant vector illustration rather than photographic rendering.

GPT-5.6 Terra

Six ivory heads and rounded lobes create a striking radial emblem around a clear hexagonal opening. Soft shading gives the bodies volume, but abrupt rectangular and diagonal band endings make the continuous alternating weave ambiguous. Repetitive scale lines and simplified faces convey an ornamental serpent more convincingly than living anatomy.

Score breakdown

CategoryGLM-5.3GPT-5.6 Terra
Task & likeness9.0/107.5/10
Anatomy & construction7.0/105.5/10
Space & placement7.0/107.0/10
Light & reflections8.0/107.5/10
Purposeful detail8.0/106.5/10
Artistry & character7.0/105.5/10
Materials & effects7.5/106.5/10
Visual integrity8.0/105.0/10

Benchmark task

GLM-5.3

An imagined creature encounters its reflection. The scene tests original character design, visual self-interpretation and the relationship between a subject and its reflected image.

GPT-5.6 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.

Mona Lisa

Same task instructions

Precise visual recall and portrait reconstruction.

The judgment

GLM-5.3

A recognizable, coherent interpretation with purposeful facial accents, readable shading and intact principal attachments. These demonstrated strengths warrant modest upward calibration. Frontal seating, distorted proportions, schematic hands and uniform surfaces still limit likeness, anatomy, depth and artistic refinement.

GPT-5.6 Terra

The dark clothing, central portrait and restrained smile evoke the Mona Lisa, but replication is limited by the broad face, elongated neck, nearly frontal pose and indistinct landscape. Soft gradients give the face some volume. Crossed arms become rounded masses without readable fingers or wrists, and the expression lacks the original's subtlety. Clean framing and curves support an otherwise basic, uneven portrait.

Score breakdown

CategoryGLM-5.3GPT-5.6 Terra
Task & likeness4.5/104.0/10
Anatomy & construction4.5/103.0/10
Space & placement4.5/104.5/10
Light & reflections6.0/105.0/10
Purposeful detail5.0/104.0/10
Artistry & character5.0/104.0/10
Materials & effects5.0/104.0/10
Visual integrity6.0/106.0/10

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

GLM-5.3

A coherent beach illustration with a clear sunset, playful subjects and purposeful small details. Varied silhouettes and broken reflections strengthen its composition. Basic human anatomy, shallow foreground construction and limited material differentiation constrain the result. The kite line reaches the smaller figure rather than requiring a grip from the nearby animal.

GPT-5.6 Terra

Warm sunset gradients, framing palms and colorful beach activities create an appealing scene. Animals, birds and a sun reflection fulfill the brief. The dog and turtle remain schematic, foreground depth is shallow, and the reflection forms rigid bands. Clean silhouettes and readable accessories support a competent illustration, but limited anatomy, material variation and lighting keep it below exceptional.

Score breakdown

CategoryGLM-5.3GPT-5.6 Terra
Task & likeness8.5/107.5/10
Anatomy & construction5.5/105.0/10
Space & placement6.0/106.0/10
Light & reflections7.0/105.0/10
Purposeful detail7.0/106.5/10
Artistry & character5.5/105.0/10
Materials & effects6.5/105.5/10
Visual integrity7.0/105.0/10

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

GLM-5.3

This benchmark has not been assessed yet.

GPT-5.6 Terra

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

GLM-5.3

A coherent sunset illustration with readable steering, expressive accessories and effective translucent water effects. Its strongest physical lighting evidence is the aligned sun reflection. The raccoon's seated support remains unresolved, while simplified joints and limited viewpoint integration constrain construction and artistry.

GPT-5.6 Terra

Expressive faces, a recognizable raccoon perched on the dog, and a sweeping wake make an appealing cartoon. The principal weakness is the steering interaction: the dog's paw sits beside the control assembly without a convincing handlebar grip. Bold outlines and coherent layering provide polish, while simplified shading, compressed seating geometry and angular splashes limit depth and physical credibility.

Score breakdown

CategoryGLM-5.3GPT-5.6 Terra
Task & likeness7.0/107.5/10
Anatomy & construction6.0/105.0/10
Space & placement6.0/107.0/10
Light & reflections7.0/106.0/10
Purposeful detail7.5/107.0/10
Artistry & character6.0/105.0/10
Materials & effects7.5/107.0/10
Visual integrity5.5/107.5/10

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

GLM-5.3

A coherent coastal illustration with a distinctive pelican silhouette, expressive flowing shapes and substantial bicycle detail. Feet meet plausible pedal platforms, but steering contact remains unresolved. Simplified leg articulation, modest depth and approximate directional lighting limit physical construction. The apparent neck opening is ambiguous rather than a verified break.

GPT-5.6 Terra

The oversized bill, expressive eye and balanced twilight palette make the cyclist immediately readable. Clean contours and wheel details support the illustration, but the rabbit-like head tufts weaken pelican likeness. The wing does not clearly grip the handlebar, pedal contact is uneven, and the drivetrain is incomplete. Lighting remains mostly decorative.

Score breakdown

CategoryGLM-5.3GPT-5.6 Terra
Task & likeness7.0/106.0/10
Anatomy & construction5.0/106.0/10
Space & placement6.0/107.0/10
Light & reflections6.0/106.5/10
Purposeful detail7.5/107.0/10
Artistry & character6.0/106.5/10
Materials & effects6.5/106.5/10
Visual integrity6.0/107.0/10

Benchmark task

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