STEALTHMARK · HEAD TO HEAD

All comparisons ↗

GPT-6 Luna vs Nemotron 3 Ultra

GPT-6 Luna leads Nemotron 3 Ultra by 22.4 points on StealthMark, 39.1 to 16.7. It leads on all 9 benchmarks. GPT-6 Luna scores higher on Newton's Mirror, with its largest category advantage in state at 600 ms.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

39.1/100Overall rank 40/58
16.7/100Overall rank 55/58

General Intelligence

31.3/100
12.2/100

Visual Intelligence

52.0/100
26.5/100

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

Cost efficiency

How much Performance you get for your money.

BENCHMARK BY BENCHMARK

Where each model leads

  • GPT-6 Luna
  • Nemotron 3 Ultra
  • Other models
  • Median

DATASHEET

Specs and pricing

GPT-6 Luna scores higher on StealthMark and is 4.8× cheaper per output token. Both take about 1.05M tokens of context.

  • GPT-6 Luna
  • Nemotron 3 Ultra
  • Other models
  • Median
PricePer 1M output tokens, input below
GPT-6 Luna: $0.504.8× cheaper$0.10 inputNemotron 3 Ultra: $2.40$0.60 input
Context windowTokens per request, max output below
GPT-6 Luna: 1.05MTied128K max outputNemotron 3 Ultra: 1MTied
Output speedTokens per second, our measurement
GPT-6 Luna: 131Tiedtokens/sNemotron 3 Ultra: 139Tiedtokens/s
SizeTotal and active parameters
GPT-6 Luna: Not disclosedNemotron 3 Ultra: 550B55B active
Developer
OpenAINVIDIA
Released
22 Sep 20264 Jun 2026
Origin
United StatesBoth
Input → output
Text, image → textText → text
Architecture
Not publishedLatentMoE: Mamba-2 + MoE + Attention hybrid with Multi-Token Prediction

Full datasheets, every price and sourceGPT-6 Luna ↗Nemotron 3 Ultra ↗

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 Luna 29.8/100Rank 37/54

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

Nemotron 3 Ultra 16.9/100Rank 48/54

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

Points per level

LevelGPT-6 LunaNemotron 3 Ultra
Level 1Second Thoughts
15 ptsOut of lives
4 ptsOut of lives
Level 2Pressure Pot
5 ptsOut of lives
4 ptsOut of lives
Level 3Hide or Flight
6 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.

GPT-6 Luna 44.0/100Rank 31/55

Cleared 4 of 5
Latch: complete playing field
Level 513 ptsTurn 15 / 40

Nemotron 3 Ultra 5.0/100Rank 49/554 of 5 levels

4 of 5 played
Latch: complete playing field
Level 413.8 ptsTurn 21 / 40

Points per level

LevelGPT-6 LunaNemotron 3 Ultra
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
8.8 ptsClearedFinished early · 1 beacon left
3.2 ptsFinished early
Level 3Relay vault
17 ptsCleared1 beacon left
7.4 ptsFinished early
Level 4Passing bay
15.8 ptsClearedFinished early · 3 beacons left
13.8 ptsFinished early
Level 5Second thoughts
13 ptsFinished early
Not played

Newton's Mirror

Same task instructions

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

The judgment

GPT-6 Luna

The schematic scene supports the paint contact order and compact paint scale. Its low, undersized main sphere and incorrectly placed wall images depart from the reference geometry. A few shrinking wall images provide limited optical organization, while absent lateral recursion and outer chains, broad sphere ribbons and isolated panel highlights leave most nested room and emitter mapping undeveloped.

Nemotron 3 Ultra

The candidate establishes a separate airborne black blob and partial crimson overlap, but leaves the paint interaction history and settled support unresolved. Its main silhouette, black placement and lamp projection depart substantially from the required camera geometry. A few faint reflection hints provide little support for the extensive mirror system, while the compact paint volumes are visibly disproportionate.

Score breakdown

CategoryGPT-6 LunaNemotron 3 Ultra
State at 600 ms22.0/252.0/25
Camera and placement3.5/153.5/15
Reflections5.0/1204.5/120
Materials and scale7.0/103.0/10
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

GPT-6 Luna

Sixfold symmetry, pale coloring and six outward-facing heads establish the motif. However, every neck floats clear of the knot. Squared outer sections replace the required semicircular lobes, and the large stepped opening lacks a regular hexagonal contour. Repeated scales and soft shading suggest skin, but translucent crossings and doubled edges weaken the living, interwoven construction.

Nemotron 3 Ultra

An endearing mechanical creature is clearly framed, with expressive amber eyes and readable segmented arms. The shell resembles layered green plates more than a deep coiled spiral. Detached-looking construction, absent working contacts and patchy rendering leave the central assembly action substantially unresolved.

Score breakdown

CategoryGPT-6 LunaNemotron 3 Ultra
Task & likeness4.5/104.0/10
Anatomy & construction3.5/104.0/10
Space & placement5.5/105.5/10
Light & reflections5.0/105.0/10
Purposeful detail5.0/104.0/10
Artistry & character5.0/103.0/10
Materials & effects5.0/104.0/10
Visual integrity4.5/104.5/10

Benchmark task

GPT-6 Luna

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.

Nemotron 3 Ultra

A hovering mechanical nautilus tests spiral construction, connected mechanical forms and contrasting metal and illuminated surfaces.

Mona Lisa

Different task instructions

Precise visual recall and portrait reconstruction.

The judgment

GPT-6 Luna

The dark veil, crossed arms, restrained smile and layered landscape clearly evoke Mona Lisa. Clean contours and muted colors create a coherent illustration. Replication is limited by enlarged, outlined eyes, simplified facial volume, broad shoulders and indistinct overlapping hands. Gradients suggest depth, but lack the original's subtle painterly transitions.

Nemotron 3 Ultra

The centered silhouette provides basic organization, but multicolored mottling and geometric overlays leave the portrait unresolved. No readable face, crossed hands, clothing or landscape establishes Mona Lisa replication. Texture and transparency offer limited visual effects without coherent anatomy, lighting or material distinctions.

Score breakdown

CategoryGPT-6 LunaNemotron 3 Ultra
Task & likeness4.5/100.5/10
Anatomy & construction4.0/100.0/10
Space & placement5.5/101.0/10
Light & reflections5.0/100.5/10
Purposeful detail5.0/100.5/10
Artistry & character4.5/101.5/10
Materials & effects5.0/100.0/10
Visual integrity6.0/100.5/10

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

GPT-6 Luna

The warm sky, layered sea and framing palms create an appealing beach composition. Animals, birds, toys and reflected sunlight fulfill the brief. Human anatomy remains schematic, and simplified figures offer limited expression. Missing shadows and the dolphin's splashless jump weaken physical grounding, while clean silhouettes and readable small details support the illustration.

Nemotron 3 Ultra

A recognizable beach illustration with appealing warm-cool contrast, readable layers and mostly clean shapes. Unresolved seat construction, schematic animals and broad geometric foliage give it an elementary finish. Weakly coordinated reflections and limited material differentiation further constrain depth and atmosphere.

Score breakdown

CategoryGPT-6 LunaNemotron 3 Ultra
Task & likeness7.5/106.5/10
Anatomy & construction5.0/103.0/10
Space & placement7.5/105.0/10
Light & reflections5.0/101.5/10
Purposeful detail7.0/104.5/10
Artistry & character5.0/103.0/10
Materials & effects6.5/102.5/10
Visual integrity6.0/106.5/10

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 Luna

This benchmark has not been assessed yet.

Nemotron 3 Ultra

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 Luna

Expressive faces, warm sunset colors and clean silhouettes create an appealing illustration. The raccoon sits on the dog's back, but the dog's grip and steering linkage remain unclear. The broad hull resembles a speedboat, and flat shading, absent sun reflections and modest water displacement limit depth and motion.

Nemotron 3 Ultra

The masked raccoon, dog and orange craft make the scene recognizable, with an appealing sunset palette. Overlapping bodies obscure the riding relationship, and the dog's connection to the controls remains unresolved. Weak water displacement, inconsistent shading and visible shape artifacts limit the finish. Ambiguous attachments do not establish detached limbs or inverted anatomy.

Score breakdown

CategoryGPT-6 LunaNemotron 3 Ultra
Task & likeness8.0/103.5/10
Anatomy & construction5.5/101.5/10
Space & placement7.5/104.5/10
Light & reflections6.0/102.0/10
Purposeful detail7.0/103.5/10
Artistry & character5.5/101.5/10
Materials & effects6.5/103.0/10
Visual integrity5.0/102.0/10

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 Luna

The pelican’s distinctive bill and the coordinated vintage poster palette make an appealing illustration. Circular wheels, orderly spokes and clean lettering strengthen the design. Riding mechanics are less convincing: neither wing reaches the handlebars, pedal contact is ambiguous, and the drivetrain is incomplete. The wheels sit against water without a clear supporting surface. Flat lighting limits volume.

Nemotron 3 Ultra

A recognizable, playful pelican dominates a readable bicycle scene. Its bill and silhouette succeed, but feet below the pedals and confused drivetrain connections weaken the riding action. Local material modeling adds interest; overlapping contours, repetitive detail and inconsistent shadows leave the illustration basic and uneven.

Score breakdown

CategoryGPT-6 LunaNemotron 3 Ultra
Task & likeness8.0/104.0/10
Anatomy & construction5.0/103.5/10
Space & placement6.5/103.5/10
Light & reflections6.5/104.0/10
Purposeful detail7.5/105.5/10
Artistry & character6.5/104.0/10
Materials & effects6.5/106.0/10
Visual integrity7.0/103.0/10

Benchmark task

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