STEALTHMARK · HEAD TO HEAD

All comparisons ↗

GPT-6 Astra vs Kimi K3.1 Pro

GPT-6 Astra leads Kimi K3.1 Pro by 18.0 points on StealthMark, 74.7 to 56.7. It leads on all 9 benchmarks. GPT-6 Astra scores higher on Newton's Mirror, with its largest category advantage in reflections.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

74.7/100Overall rank 1/53
56.7/100Overall rank 11/53

General Intelligence

73.4/100
53.9/100

Visual Intelligence

77.0/100
62.0/100

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

  • GPT-6 Astra

BENCHMARK BY BENCHMARK

Where each model leads

  • GPT-6 Astra
  • Kimi K3.1 Pro
  • Other models
  • Median

DATASHEET

Specs and pricing

GPT-6 Astra scores higher on StealthMark. Kimi K3.1 Pro is 5× cheaper per output token and runs 1.8× faster in our speed test. Both take about 1.05M tokens of context.

  • GPT-6 Astra
  • Kimi K3.1 Pro
  • Other models
  • Median
PricePer 1M output tokens, input below
GPT-6 Astra: $50$10 inputKimi K3.1 Pro: $105× cheaper$2 input
Context windowTokens per request, max output below
GPT-6 Astra: 1.05MTied128K max outputKimi K3.1 Pro: 1MTied1M max output
Output speedTokens per second, our measurement
GPT-6 Astra: 65tokens/sKimi K3.1 Pro: 1171.8× fastertokens/s
SizeTotal and active parameters
GPT-6 Astra: Not disclosedKimi K3.1 Pro: 2.8T104B activeEstimated
Developer
OpenAIMoonshot AI
Released
3 Sep 202623 Oct 2026
Origin
United StatesChina
Input → output
Text, image → textText, image, video → text
Architecture
Not publishedPost-trained version of Kimi K3, a Mixture-of-Experts model

Full datasheets, every price and sourceGPT-6 Astra ↗Kimi K3.1 Pro ↗

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 Astra 76.1/100Rank 2/47

Survived 3 of 3
Foresight: complete playing field
Level 318 ptsTurn 20 / 20

Kimi K3.1 Pro 61.1/100Rank 10/47

Survived 1 of 3
Foresight: complete playing field
Level 315 ptsTurn 20 / 20

Points per level

LevelGPT-6 AstraKimi K3.1 Pro
Level 1Second Thoughts
27 ptsSurvived
15 ptsOut of lives
Level 2Pressure Pot
11 ptsSurvived
12 ptsOut of lives
Level 3Hide or Flight
18 ptsSurvived
15 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.

GPT-6 Astra 87.6/100Rank 1/48

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

Kimi K3.1 Pro 81.8/100Rank 8/48

Cleared 5 of 5
Latch: complete playing field
Level 519.8 ptsTurn 31 / 40

Points per level

LevelGPT-6 AstraKimi K3.1 Pro
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
10.2 ptsClearedFinished early · 1 beacon left
9.8 ptsCleared
Level 3Relay vault
19.2 ptsCleared
19.2 ptsCleared
Level 4Passing bay
18.2 ptsClearedFinished early · 3 beacons left
18.2 ptsClearedFinished early · 3 beacons left
Level 5Second thoughts
20.6 ptsCleared
19.8 ptsCleared

Newton's Mirror

Same task instructions

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

The judgment

GPT-6 Astra

Major placement and the visible contact sequence are convincing. Dense planar light repetition, recognizable spherical panel images and simplified nested sphere motifs provide useful optical structure. Missing outer chains, truncated distant sequences and incorrect bead-like belts still leave the recursive reflections substantially incomplete.

Kimi K3.1 Pro

The settled sphere, separate airborne black paint and crimson-before-cyan interaction are supported, with a modest crimson travel overshoot. A shared downward framing displacement remains after accounting for the square canvas. The principal limitation is the reflection system: sparse sphere copies and generic curved bands omit most of the ordered mirror-room structure. Materials receive a small upward revision because bounded bright images and dark bands visibly establish some polish, while soft shading and bulky paint remain separate weaknesses.

Score breakdown

CategoryGPT-6 AstraKimi K3.1 Pro
State at 600 ms25.0/2523.5/25
Camera and placement15.0/1510.5/15
Reflections70.0/12026.0/120
Materials and scale10.0/107.0/10
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 Astra

The sixfold silhouette, pale coloring, outward heads and central hexagon closely match the brief. Fine scales and soft shading support rounded volume. However, several crossings have conspicuous rectangular seams and weak contact shadows, making the weave feel assembled. Small heads have limited anatomical definition, and the repetitive texture conveys less organic life than requested.

Kimi K3.1 Pro

The transformed reflection, attentive pose and constellation anatomy clearly communicate introspection. Purposeful fine details enrich a coherent composition. Tubular limbs, crowded lower-body overlaps, shallow depth and limited cast lighting keep the overall execution within the competent range.

Score breakdown

CategoryGPT-6 AstraKimi K3.1 Pro
Task & likeness9.0/108.0/10
Anatomy & construction8.0/105.5/10
Space & placement7.5/106.5/10
Light & reflections7.0/106.5/10
Purposeful detail8.0/107.5/10
Artistry & character8.5/107.0/10
Materials & effects8.0/107.0/10
Visual integrity8.5/107.0/10

Benchmark task

GPT-6 Astra

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.

Kimi K3.1 Pro

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.

Mona Lisa

Same task instructions

Precise visual recall and portrait reconstruction.

The judgment

GPT-6 Astra

The crossed hands, restrained smile, dark clothing and hazy landscape clearly recall the Mona Lisa. Replication is limited by the frontal, elongated face, schematic eyes and stiff fingers. Soft shading gives basic volume, but broad garment shapes and angular scenery lack the original’s nuanced depth and painterly transitions. The frame and composition are controlled; surface grain adds texture without resolving the simplified drawing.

Kimi K3.1 Pro

A coherent, recognizable interpretation with controlled color and readable landscape layers. Replication is limited by the frontal seated construction, elongated facial proportions and oversized, weakly articulated hands. Smooth shading supplies some volume but little of the reference's subtle expression, fabric structure or atmospheric transitions.

Score breakdown

CategoryGPT-6 AstraKimi K3.1 Pro
Task & likeness6.5/105.0/10
Anatomy & construction6.5/105.0/10
Space & placement6.5/105.5/10
Light & reflections6.5/106.0/10
Purposeful detail6.5/105.5/10
Artistry & character7.0/106.0/10
Materials & effects7.0/105.0/10
Visual integrity7.5/106.5/10

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 Astra

A richly composed sunset beach illustration with cohesive peach, mauve and teal colors. Sweeping palm fronds, a scalloped umbrella and curved shorelines frame the sun and its aligned reflection. Distinct animal silhouettes and carefully distributed beach details reward inspection. Simplified human and dog anatomy, slight dolphin splash cues and inconsistent equipment shadows limit physical conviction.

Kimi K3.1 Pro

A coherent, playful beach illustration with readable animals, an organized composition and an aligned sunset reflection. Connected silhouettes, sand texture and atmospheric effects demonstrate controlled execution. Simplified joints, flattened foreground viewpoints and inconsistent shadows limit anatomical and spatial credibility.

Score breakdown

CategoryGPT-6 AstraKimi K3.1 Pro
Task & likeness9.0/108.5/10
Anatomy & construction7.5/105.5/10
Space & placement8.5/105.5/10
Light & reflections7.5/106.5/10
Purposeful detail8.5/106.5/10
Artistry & character8.5/105.5/10
Materials & effects7.5/106.5/10
Visual integrity7.5/107.5/10

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 Astra

This benchmark has not been assessed yet.

Kimi K3.1 Pro

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 Astra

Expressive animals, a balanced sunset composition and sweeping wake make an appealing illustration. The raccoon rides the dog's back, and the dog's connected foreleg reaches the controls. The compact steering assembly and layered paws are less clearly constructed. Lighting remains mostly flat, with little shadow or sun reflection. Clean contours and purposeful details support strong illustrative execution.

Kimi K3.1 Pro

The additional comparisons support retaining the assessment. Readable steering, shaped animal contours and purposeful details coexist with simplified joints, a frontal passenger whose seated weight is weakly conveyed, and limited physical lighting. The angled craft establishes more depth than the riding pose resolves.

Score breakdown

CategoryGPT-6 AstraKimi K3.1 Pro
Task & likeness8.5/107.0/10
Anatomy & construction7.5/105.5/10
Space & placement8.0/106.0/10
Light & reflections6.5/105.5/10
Purposeful detail8.0/107.0/10
Artistry & character7.5/106.0/10
Materials & effects8.0/106.0/10
Visual integrity8.5/107.0/10

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 Astra

The pelican's long bill, flowing scarf and warm coastal palette create a charming, balanced illustration. Wheels, spokes and frame are largely coherent, but neither wing reaches the handlebars, and the rear leg's pedal contact is unclear. Flat lighting and simplified surfaces limit depth despite clean outlines and attractive feather detail.

Kimi K3.1 Pro

A characterful coastal illustration with controlled contours and varied feather, accessory and bicycle detail. Layered scenery and subtle shading enrich the presentation. Soft leg articulation, unresolved riding contacts, shallow depth and generalized lighting limit physical precision.

Score breakdown

CategoryGPT-6 AstraKimi K3.1 Pro
Task & likeness8.0/108.5/10
Anatomy & construction7.0/107.5/10
Space & placement7.5/106.5/10
Light & reflections7.0/106.5/10
Purposeful detail8.0/107.5/10
Artistry & character8.0/107.5/10
Materials & effects7.5/106.5/10
Visual integrity8.0/107.5/10

Benchmark task

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