Atlas Graph Explorer

model:claude-opus-4-6@current

ModelVersioncompute/models/claude-opus-4-6.yaml·Open in Graph →

overview json graph

Attributes

modelFamilyId

model-family:claude-opus-4

versionRange

>=4.6.0 <4.7.0

kind

chat

contextWindowTokens

500000

costPerMTokInput

costPerMTokOutput

supportsExtendedThinking

true

supportsAdaptiveThinking

false

thinkingEffortLevels

low
medium
high

supportsThinkingBudgetTokens

true

modalities

text
image

modalityDirections

kind
text
direction
input
kind
image
direction
input
kind
text
direction
output

releaseDate

2025-10-15

lifecycleStatus

lifecycleStatusChangedAt

2025-10-15

knowledgeCutoffDate

2025-04-01

safetyTier

anthropic-asl-3

modelCardUrl

https://www.anthropic.com/claude/opus

supportsFineTuning

false

regions

global

Outgoing edges (15)

realizes1

layer:1-model·LayerModel

served_by3

provider:anthropic·ModelProviderProductAnthropic
provider:aws-bedrock·ProviderAWS Bedrock
provider:gcp-vertex·ProviderGoogle Cloud Vertex AI

speaks1

model-transport:anthropic-messages·ModelTransportProtocolAnthropic Messages

supports9

capability:supports-batch-api·CapabilitySupports batch API
capability:extended-thinking·CapabilityExtended thinking
capability:parallel-tool-calls·CapabilityParallel tool calls
capability:supports-prompt-caching·CapabilitySupports prompt caching
capability:streaming·CapabilityStreaming
capability:supports-text-streaming·CapabilitySupports text streaming
capability:supports-thinking-budget-tokens·CapabilitySupports thinking budget tokens
capability:supports-thinking-streaming·CapabilitySupports thinking streaming
capability:supports-tool-call-streaming·CapabilitySupports tool-call streaming

version_of1

model-family:claude-opus-4·ModelFamilyClaude Opus 4

Incoming edges (22)

about_subject16

has_version1

model-family:claude-opus-4·ModelFamilyClaude Opus 4

serves2

provider:anthropic·ModelProviderProductAnthropic
provider:gcp-vertex·ProviderGoogle Cloud Vertex AI

supersedes1

model:claude-opus-4-7@current·ModelVersion

supported_by1

capability:extended-thinking·CapabilityExtended thinking

targets1

benchmark:gpqa·BenchmarkGPQA