The Orchestrator › Models › Anthropic

Anthropic: Claude Sonnet 5

Sonnet 5 is Anthropic's most capable Sonnet-class model, with frontier performance across coding, agents, and professional work. It supports adaptive thinking with selectable reasoning effort levels (low, medium, high, max,...

Specifications

ProviderAnthropic
Model IDanthropic/claude-sonnet-5
Context window1M tokens (1,000,000)
Modalitiestext + image + file → text
TokenizerClaude
Router sourcesOpenRouter, TrustedRouter, Vercel AI Gateway, Cloudflare
Released2026-06-30
Tracked since2026-11-14
Popularity#35 by usage · peak 3.7% spend share in an OpenRouter category
ReasoningOn by default

Pricing (per million tokens)

VariantSourceAPI IDInputOutputContext
standardOpenRouteranthropic/claude-sonnet-5$2.00$10.001M
batchOpenRouteranthropic/claude-sonnet-5:batch$1.00$5.001M
standardTrustedRouteranthropic/claude-sonnet-5$2.11$10.551M
awsTrustedRouteranthropic/claude-sonnet-5:aws$2.11$10.551M
standardVercel AI Gatewayanthropic/claude-sonnet-5$2.00$10.001M
batchVercel AI Gatewayanthropic/claude-sonnet-5:batch$1.00$5.001M
standardCloudflareanthropic/claude-sonnet-5$2.00$10.00—

Recorded price history

DateInput / 1MOutput / 1MContext
2026-11-14$2.00$10.00—
2026-11-14$2.00$10.001M
2026-11-14$2.11$10.551M
2026-11-14$2.00$10.001M

Capabilities

Vision (image input)File inputAgentic tool useStructured outputsReasoning mode1M+ context windowControllable reasoningNo content storage (attested)Zero data retentionNo training on inputs (attested)

Trust flags come from TrustedRouter's per-provider attestations.

Supported parameters

include_reasoning, max_completion_tokens, max_tokens, reasoning, reasoning_effort, response_format, stop, structured_outputs, tool_choice, tools, verbosity