← All models

Claude Opus 5

Anthropic

flagship

Capabilities

VisionToolsStructured OutputExtended ThinkingLong ContextCode

Strengths

  • Exceptional reasoning and multi-step problem-solving
  • Best-in-class code generation and analysis
  • Strong vision understanding with 1M token context
  • Extended thinking for complex tasks
  • Reliable tool use and structured output

Weaknesses

  • Extremely slow inference (0 tok/s) — unsuitable for real-time applications
  • High output pricing ($25/M tokens) limits cost-sensitive deployments
  • Closed-source with limited transparency on training and methodology
  • No self-hosting option due to proprietary weights

Pricing

Input / 1M tokens

$5

Output / 1M tokens

$25

Context window

1,000k

Transparency

Open weights

0.0 / 10

Open training data

0.0 / 10

Open methodology

3.5 / 10

Licence openness

1.0 / 10

Provider disclosure

3.0 / 10

FMTI company score

N/A

Composite:4.0 / 10

Anthropic publishes research on constitutional AI and safety practices but does not disclose model weights, training data, or detailed methodology. Limited transparency compared to open-weight alternatives.

Sustainability

Inference energy

N/A

Training footprint

N/A

Provider infrastructure

5.5 / 10

Composite:2.0 / 10

No public commitments to renewable energy or detailed sustainability reporting. Flagship models typically have high inference costs; sustainability data unavailable.

Task Fitness

question answering
99%
code
97%
mathematics
50%
business communication
50%
vision
88%
analysis
99%
extraction
72%
long-form generation
92%
research
50%
reasoning
98%
summarisation
95%
translation
90%
conversation
93%