STEALTHMARK · HEAD TO HEAD

All comparisons ↗

GPT-6.1 Terra vs Ling 3.1 Flash

GPT-6.1 Terra leads Ling 3.1 Flash by 18.2 points on StealthMark, 60.4 to 42.2. It leads on 7 of 9 benchmarks, including both games, and Ling 3.1 Flash on 2. GPT-6.1 Terra scores higher on Ground Truth.

See the difference ↘
All results ↗
All results ↗

Overall StealthMark

60.4/100Overall rank 9/54
42.2/100Overall rank 33/54

General Intelligence

58.0/100
35.5/100

Visual Intelligence

64.9/100
51.2/100

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

Open or local

Open modelMIT

BENCHMARK BY BENCHMARK

Where each model leads

  • GPT-6.1 Terra
  • Ling 3.1 Flash
  • Other models
  • Median

DATASHEET

Specs and pricing

GPT-6.1 Terra scores higher on StealthMark and takes 4× more context. Ling 3.1 Flash is 5.6× cheaper per output token.

  • GPT-6.1 Terra
  • Ling 3.1 Flash
  • Other models
  • Median
PricePer 1M output tokens, input below
GPT-6.1 Terra: $5$1 inputLing 3.1 Flash: $0.905.6× cheaper$0.30 input
Context windowTokens per request, max output below
GPT-6.1 Terra: 1.05M4× larger128K max outputLing 3.1 Flash: 256K
Output speedTokens per second, our measurement
GPT-6.1 Terra: Not measuredLing 3.1 Flash: 115tokens/s
SizeTotal and active parameters
GPT-6.1 Terra: Not disclosedLing 3.1 Flash: 560B25B active
Developer
OpenAIInclusionAI
Released
26 Oct 202629 Sep 2026
Origin
United StatesChina
Input → output
Text, image → textText → text
Architecture
Not publishedHybrid-linear MoE; 72 layers (63 KDA, 9 gated MLA); 512 routed experts, 8 selected and 1 shared.

Full datasheets, every price and sourceGPT-6.1 Terra ↗Ling 3.1 Flash ↗

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

Ling 3.1 Flash 43.1/100Rank 45/501 of 3 levels

1 of 3 played
Foresight: complete playing field
Level 26 ptsTurn 14 / 18

Points per level

LevelGPT-6.1 TerraLing 3.1 Flash
Level 1Second Thoughts
21 ptsSurvived
Pending
Level 2Pressure Pot
9 ptsOut of lives
6 ptsOut of lives
Level 3Hide or Flight
15 ptsSurvived
Not played

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

Ling 3.1 Flash 45.9/100Rank 39/513 of 5 levels

3 of 5 played
Latch: complete playing field
Level 49.8 ptsTurn 21 / 40

Points per level

LevelGPT-6.1 TerraLing 3.1 Flash
Level 1Hold the door
5.2 ptsCleared
5.2 ptsCleared
Level 2Crossfeed
9.8 ptsCleared
9.4 ptsCleared
Level 3Relay vault
19.2 ptsCleared
Pending
Level 4Passing bay
17.8 ptsClearedFinished early · 3 beacons left
9.8 ptsFinished early
Level 5Second thoughts
20.6 ptsCleared
Not played

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.

Ling 3.1 Flash

Central placement, separate airborne black paint and the elongated crimson coating earn substantial state and placement credit. Cyan contact remains unresolved. A few isolated sphere images provide limited optical evidence, while the defining repeated mirror structure is missing. Readable opaque metallic volume and restrained paint scale justify increased material credit. Core: 61.5/170; including the floor reward: 61.5/180.

Score breakdown

CategoryGPT-6.1 TerraLing 3.1 Flash
State at 600 ms23.0/2522.0/25
Camera and placement11.0/1513.5/15
Reflections33.0/12018.0/120
Materials and scale7.5/107.0/10
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-6.1 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.

Ling 3.1 Flash

An appealing, coherent storybook ant with a clear expression, complete silhouette and readable signal mechanism. Clean curves and warm accents support the composition, while simplified joints, schematic hardware, limited material differentiation and modest light interaction constrain its finish.

Score breakdown

CategoryGPT-6.1 TerraLing 3.1 Flash
Task & likeness7.5/107.5/10
Anatomy & construction6.5/107.5/10
Space & placement6.5/107.0/10
Light & reflections6.0/107.5/10
Purposeful detail6.5/107.0/10
Artistry & character6.5/107.0/10
Materials & effects6.0/106.5/10
Visual integrity6.0/108.0/10

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.

Ling 3.1 Flash

An ant tending signals in a miniature workshop tests insect anatomy, purposeful gestures and the placement of small objects in a lit environment.

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.

Ling 3.1 Flash

A basic, uneven Mona Lisa interpretation with a clear facial focal point, articulated eyes and modest tonal volume. Weak likeness, oversized head proportions, unclear shoulders, curled bare arms and indistinct hands limit replication. The compressed composition and rectangular torso weaken an otherwise coherent earthy palette.

Score breakdown

CategoryGPT-6.1 TerraLing 3.1 Flash
Task & likeness5.5/102.0/10
Anatomy & construction6.5/101.5/10
Space & placement4.5/104.0/10
Light & reflections6.5/103.5/10
Purposeful detail6.0/103.5/10
Artistry & character6.0/102.0/10
Materials & effects6.0/101.5/10
Visual integrity6.5/102.0/10

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.

Ling 3.1 Flash

A coherent sunset illustration with harmonious color, clean silhouettes and readable beach details. Its calm composition works well, while playful activity is modest. Simplified animal anatomy, shallow shoreline layering, repetitive foliage and a geometric reflection limit drawing sophistication and naturalism.

Score breakdown

CategoryGPT-6.1 TerraLing 3.1 Flash
Task & likeness8.5/107.5/10
Anatomy & construction7.5/106.0/10
Space & placement6.5/107.0/10
Light & reflections5.5/107.0/10
Purposeful detail7.5/106.5/10
Artistry & character7.0/106.5/10
Materials & effects6.5/106.5/10
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.

Ling 3.1 Flash

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.

Ling 3.1 Flash

An appealing sunset cartoon with recognizable animals, expressive faces and coherent landscape layers. The sky and warm water streaks provide its strongest rendering. Ambiguous seated support, stiff forelegs, a shallow hull and an unclear steering connection limit construction. Clean shapes and readable accessories add finish, while restrained spray and flat surfaces limit material and motion effects.

Score breakdown

CategoryGPT-6.1 TerraLing 3.1 Flash
Task & likeness6.5/107.5/10
Anatomy & construction5.5/105.5/10
Space & placement6.0/107.0/10
Light & reflections5.0/107.0/10
Purposeful detail7.0/107.0/10
Artistry & character6.0/105.5/10
Materials & effects6.0/105.5/10
Visual integrity7.0/106.0/10

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.

Ling 3.1 Flash

A clean, appealing pelican illustration with a readable bicycle and coherent landscape. Bill shading, feather layers and faint wheel shadows provide modest depth. Uncertain pedal contact, simplified leg construction and limited material rendering keep the result within competent illustration quality.

Score breakdown

CategoryGPT-6.1 TerraLing 3.1 Flash
Task & likeness7.0/107.5/10
Anatomy & construction7.0/105.5/10
Space & placement6.0/107.0/10
Light & reflections5.5/106.5/10
Purposeful detail7.0/106.0/10
Artistry & character6.5/106.5/10
Materials & effects5.5/106.5/10
Visual integrity6.0/106.0/10

Benchmark task

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