stealthmodels.

MODEL FIELD GUIDE

Claude Opus 5.5: Pelican

Anthropic's top Opus tier focuses on complex reasoning and coding, with a million-token context for large tasks.

Drawn by this model01 / 01
Pelican benchmark result generated by this model
SVG BENCHMARKPelicanView result

A good fit for

Long-running coding agents and demanding professional analysis where resolving the task matters more than minimizing token cost.

Context window
1,000,000tokens[1]
Model factsArchitecture, context & capabilities
First available
2026-09-22[1]
Developer
Anthropic[1]
Origin
United States[2]
Family
Claude 5.5[1]
Context window
1,000,000 tokens[1]
Maximum output
128,000 tokens[1]
Input
text, image[1]
Output
text[1]
API access & pricingOpenRouter · $4 input / $20 output · per 1M tokens

Anthropic API; input 4 USD/1M; cached read 0.2/1M; output 20/1M; 5-minute cache write $5/M; 1-hour cache write $8/M. Batch rates shown as 50% of standard.

[1]
  • OpenRouterObserved 2026-10-06
    Input $4Output $20Cached read $0.2Cached write $5

    per 1M tokens

    anthropic/claude-opus-5.5[3]
  • Anthropic APIObserved 2026-10-06
    Input $4Output $20Cached read $0.2Cached write $5Batch input $2Batch output $10

    per 1M tokens

    Available

    claude-opus-5-5[1]
Measurements & sources4 primary references · Speed & serving conditions
  • AnthropicObserved 2026-10-06
    [1]
  • Artificial AnalysisObserved 2026-10-06

    94 tokens/s decode

    release comparison

    [4]

StealthMark score

StealthMark measures fluid and visual intelligence of AI models.

General and Visual

General Intelligence

Perspective, anatomy, correct shadows, purposeful detail and physical consistency.

Visual Intelligence

Likeness, anatomy, detail, light, materials, composition and artistic expression.

Judging

Frontier Models blindly judge the AI generation unbiased through an intricate process. Our researchers control for mistakes.

Overall score

Overall combines weighted task scores on a scale from 0 to 100. Repeating a task does not increase its weight. Estimated means the total includes an estimate for one missing scene.

Benchmaxxing

Benchmaxx flags models that score unusually well on the familiar Pelican task compared with their other benchmark results. The percentage measures that imbalance, not the probability of training contamination. Flagged Pelican scores are excluded from Overall. Pelican never counts toward General or Visual.

We keep some prompts private to discourage test-specific optimization.

Scoring categories

Prompt & likeness

Subjects, actions and likeness.

Anatomy & construction

Coherent bodies, joints and machinery.

Space & placement

Perspective, scale and contact.

Light & reflections

Lighting, shadows and reflections.

Purposeful detail

Clear, purposeful small features.

Artistry & character

Composition, expression and drawing skill.

Materials & effects

Surfaces, texture and movement.

Visual integrity

Clean shapes, layers and edges.

BENCHMARK RESULTS

Compare results

This link opens the same results in the same order.

StealthMark benchmark results

Reset benchmark preferences?

Return to the default view, clear comparison selections and restore the advertisement.