AnthropicAnthropic·text

Claude Opus 4.7

CodeVisionReasoningWeb searchFunction calling
Advertised as claude-opus-4-7claude-opus-4.7opus-4-7opus-4.7
Quick reference
Claude Opus 4.7 — TLDR
  • 🏢 Anthropic flagship Opus model, released April 2026.
  • 📏 1M-token context window with 128K max output tokens.
  • 🧠 Adaptive thinking dynamically allocates reasoning budget per task.
  • 🔧 Built for long-horizon agentic coding and enterprise workflows.
  • 👁️ High-resolution vision for charts, diagrams, dense documents.
  • 🎯 New "xhigh" effort level between high and max.
  • ⚡ Reported CursorBench above 70%, up from 58% prior.
  • 🆕 Succeeded by Claude Opus 4.8 in May 2026.
💰 Best price on AntSeed
$0.200 / $1.0096%
per 1M · cheapest in / out
📏 Context
1M tokens
🐜 Sellers
25
advertising on AntSeed
Provider

Anthropic PBC is an American artificial intelligence company headquartered in San Francisco. Structured as a public benefit corporation, the lab develops large language models under the Claude name, with a research emphasis on building reliable, steerable, and safety-focused AI…

Explore 12 more models by Anthropic
About this model

Claude Opus 4.7 is Anthropic's most capable generally available model at its April 2026 launch, positioned for complex reasoning and long-running agentic coding. It carries a 1M-token context window, 128K max output tokens, and adaptive thinking—an upgrade to extended thinking that lets the model decide how much to reason based on task complexity and a selected effort level. It also adds high-resolution multimodal vision, improving accuracy on charts, dense documents, and screen UIs.

Compared with its predecessor [[sibling:claude-opus-4-6|Claude Opus 4.6]], Anthropic describes a step-change in agentic coding and long-horizon autonomy. On Anthropic's reported CursorBench, Opus 4.7 clears 70% versus 58% for Opus 4.6, and the company cites a 14% gain on complex multi-step workflows at fewer tokens with roughly a third of the tool errors. It also introduces a new "xhigh" effort level between high and max.

The Opus 4.7 generation includes a low-latency variant, [[sibling:claude-opus-4-7-fast|Claude Opus 4.7 Fast]], alongside the standard model. Anthropic later released [[sibling:claude-opus-4-8|Claude Opus 4.8]], which builds directly on Opus 4.7 and, per Anthropic, reduces wasted thinking tokens on mixed workloads and improves tool triggering and post-compaction long-context quality.

Opus 4.7 targets autonomous agents, multi-step coding, and vision-heavy pipelines, sitting between Opus 4.6 and Opus 4.8 in Anthropic's lineup.

Sources
platform.claude.comModels overview - Claude API Docs· platform.claude.comanthropic.comIntroducing Claude Opus 4.7· anthropic.com

This About section is AI-generated from public sources via VeniceStats + Venice inference, with no human editing. It may contain inaccuracies.

Usage on AntSeed
Tokens served
326.46M
input + output
Requests
116,379
settled calls
Buyers
31
distinct, on this model
Sellers used
16
of 25 advertising
Settled
$2.00k
gross USDC, this model
Sellers serving Claude Opus 4.7 (25)compare on the network explorer →
SellerReputationRoutingInput $/MCached $/MOutput $/MCategoriesAPI

"Best price" and the seller table are live AntSeed catalog data (advertised $/1M tokens — or $ per generated image for unit-billed image models — not settled amounts). Reputation = on-chain trust (0-100). "Routing" = the SDK's default buyer routing (what the VPR desktop app ships with): a trust ≥ 60 gate on the effective reputation, then cheapest-first among routable sellers; live failover state (per-peer cooldowns) is buyer-side runtime and not included. Model knowledge (TLDR, provider, About) via the VeniceStats enrichment layer. Advertised catalog, not the model used in any specific purchase. "Usage on AntSeed" counts only settlements whose buyers share the per-model split on-chain (metadata v2/v3, opt-in), so every usage figure is a lower bound.