STEALTHMARK · HEAD TO HEAD

All comparisons ↗

GPT-6 Luna vs MiniMax M3

GPT-6 Luna leads MiniMax M3 by 0.5 points on StealthMark, 39.1 to 38.6. MiniMax M3 leads on 5 of 9 benchmarks and GPT-6 Luna on 4, including both games. MiniMax M3 scores higher on Newton's Mirror, with its largest category advantage in reflections.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

39.1/100Overall rank 40/58
38.6/100Overall rank 42/58

General Intelligence

31.3/100
29.5/100

Visual Intelligence

52.0/100
53.2/100

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

  • MiniMax M3

Cost efficiency

How much Performance you get for your money.

BENCHMARK BY BENCHMARK

Where each model leads

  • GPT-6 Luna
  • MiniMax M3
  • Other models
  • Median

DATASHEET

Specs and pricing

GPT-6 Luna scores higher on StealthMark and is 2.4× cheaper per output token. MiniMax M3 runs 23% faster in our speed test. Both take about 1.05M tokens of context.

  • GPT-6 Luna
  • MiniMax M3
  • Other models
  • Median
PricePer 1M output tokens, input below
GPT-6 Luna: $0.502.4× cheaper$0.10 inputMiniMax M3: $1.20$0.30 input
Context windowTokens per request, max output below
GPT-6 Luna: 1.05MTied128K max outputMiniMax M3: 1MTied512K max output
Output speedTokens per second, our measurement
GPT-6 Luna: 131tokens/sMiniMax M3: 16223% fastertokens/s
SizeTotal and active parameters
GPT-6 Luna: Not disclosedMiniMax M3: 428B23B active
Developer
OpenAIMiniMax
Released
22 Sep 20261 Jun 2026
Origin
United StatesSingapore
Input → output
Text, image → textText, image, video → text
Architecture
Not publishedMixture-of-Experts with MiniMax Sparse Attention and native multimodal encoder

Full datasheets, every price and sourceGPT-6 Luna ↗MiniMax M3 ↗

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

MiniMax M3 22.1/100Rank 44/54

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

Points per level

LevelGPT-6 LunaMiniMax M3
Level 1Second Thoughts
15 ptsOut of lives
17 ptsOut of lives
Level 2Pressure Pot
5 ptsOut of lives
3 ptsOut of lives
Level 3Hide or Flight
6 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.

GPT-6 Luna 44.0/100Rank 31/55

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

MiniMax M3 31.4/100Rank 41/55

Cleared 2 of 5
Latch: complete playing field
Level 58.8 ptsTurn 11 / 40

Points per level

LevelGPT-6 LunaMiniMax M3
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
8.8 ptsClearedFinished early · 1 beacon left
9.4 ptsClearedFinished early · 1 beacon left
Level 3Relay vault
17 ptsCleared1 beacon left
0.2 ptsFinished early
Level 4Passing bay
15.8 ptsClearedFinished early · 3 beacons left
14.2 ptsFinished early
Level 5Second thoughts
13 ptsFinished early
8.8 ptsFinished early

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.

MiniMax M3

The centered sphere, aligned left wall image, smaller lower-left continuation and partially occluded floor image provide recognizable local mirror coherence. Cyan remains airborne, the main sphere resembles white marble, and missing right-side imagery, recursive optics and panel mappings leave the scene substantially incomplete.

Score breakdown

CategoryGPT-6 LunaMiniMax M3
State at 600 ms22.0/257.0/25
Camera and placement3.5/1514.5/15
Reflections5.0/12022.0/120
Materials and scale7.0/105.0/10
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.

MiniMax M3

The oversized coral M silhouette dominates clearly, with an expressive charcoal face and appealing miniature body. Limbs connect coherently and the cube reads clearly. The lowered left hand only loosely suggests the requested hip pose. Thin highlights and simple gradients provide limited sculptural volume; the cube's glow and grounding shadow are restrained.

Score breakdown

CategoryGPT-6 LunaMiniMax M3
Task & likeness4.5/108.0/10
Anatomy & construction3.5/106.0/10
Space & placement5.5/107.0/10
Light & reflections5.0/106.0/10
Purposeful detail5.0/104.0/10
Artistry & character5.0/104.0/10
Materials & effects5.0/104.0/10
Visual integrity4.5/108.0/10

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.

MiniMax M3

A small robot with an oversized angular head presents a glowing object. The scene tests character proportions, gesture and contrasting materials.

Mona Lisa

Same 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.

MiniMax M3

A coherent illustrated portrait with developed facial shading, layered hair and purposeful costume detail. Replication remains limited by the frontal torso, separated hands, altered facial proportions and simplified landscape. Surface treatment and intact attachments deserve credit independently of the weaker likeness and anatomy.

Score breakdown

CategoryGPT-6 LunaMiniMax M3
Task & likeness4.5/103.5/10
Anatomy & construction4.0/104.5/10
Space & placement5.5/103.5/10
Light & reflections5.0/105.5/10
Purposeful detail5.0/106.0/10
Artistry & character4.5/105.0/10
Materials & effects5.0/106.0/10
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

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.

MiniMax M3

A cheerful, coherent sunset illustration with framing palms, warm gradients and an aligned sun reflection. Animals and beach toys add character, but their construction remains basic. The sailboat appears suspended above the water horizon, while absent cast shadows weaken foreground contact. Clean geometric shapes provide polish without substantial texture or depth.

Score breakdown

CategoryGPT-6 LunaMiniMax M3
Task & likeness7.5/108.0/10
Anatomy & construction5.0/104.5/10
Space & placement7.5/106.0/10
Light & reflections5.0/107.0/10
Purposeful detail7.0/106.0/10
Artistry & character5.0/105.0/10
Materials & effects6.5/105.0/10
Visual integrity6.0/105.5/10

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.

MiniMax M3

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.

MiniMax M3

The raccoon sits above the dog's back, and the dog reaches toward the jet ski controls. Expressive faces and a vivid sunset make the scene appealing. Construction is simplified: the dog's paws and steering grip are ambiguous, the hull is shallow, and blurred wake patches give limited water displacement. Sunset glow exceeds the lighting modeled on the riders.

Score breakdown

CategoryGPT-6 LunaMiniMax M3
Task & likeness8.0/108.0/10
Anatomy & construction5.5/105.5/10
Space & placement7.5/106.5/10
Light & reflections6.0/106.0/10
Purposeful detail7.0/107.0/10
Artistry & character5.5/106.0/10
Materials & effects6.5/106.0/10
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

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.

MiniMax M3

The pelican's long bill, pouch and curved neck read clearly, and the warm sunset gives the composition appeal. Bicycle construction weakens the riding action: feet do not convincingly engage the pedals, crank connections are unclear, and body support is unresolved. Clean silhouettes and orderly spokes provide a competent illustrative finish.

Score breakdown

CategoryGPT-6 LunaMiniMax M3
Task & likeness8.0/108.0/10
Anatomy & construction5.0/105.5/10
Space & placement6.5/106.5/10
Light & reflections6.5/106.5/10
Purposeful detail7.5/107.5/10
Artistry & character6.5/106.5/10
Materials & effects6.5/106.0/10
Visual integrity7.0/106.5/10

Benchmark task

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