claude-opus-5-5 — API, Pricing & Context Window | Vivgrid

claude-opus-5-5 is the latest model in Anthropic's Opus tier, built for long-running agentic coding and knowledge work. It keeps the 1M-token context window and 128K-token max output of claude-opus-5, runs with always-on adaptive thinking, and costs less per token than the model it succeeds.

On Vivgrid, claude-opus-5-5 is served through the native Messages API (/messages) with geo-distributed acceleration across AMER, EMEA, and APAC. It uses the same unified Vivgrid API key as the rest of the catalog, so moving an agent from claude-opus-5 is a model-string change.

Specifications

Pricing

Pricing in USD per 1M tokens, matching the provider's rates.

Quick start

Call claude-opus-5-5 through Vivgrid's unified, OpenAI-compatible endpoint. Get an API key from the Vivgrid Console.

curl https://api.vivgrid.com/v1/messages \
  -H "Authorization: Bearer $VIVGRID_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-opus-5-5",
    "max_tokens": 1024,
    "messages": [
      { "role": "user", "content": "Say hello in English, Chinese and Spanish." }
    ],
    "stream": true
  }'

Prompt caching is billed at $5.00 per 1M cache-write tokens on the 5-minute TTL and $8.00 per 1M on the 1-hour TTL, with cached reads at $0.20.

Ideal use cases

• Long-running coding agents and autonomous software development
• Whole-codebase analysis, migration, and large refactoring projects
• Knowledge work that chains many tool calls across a long session
• Teams on claude-opus-5 looking for the same context budget at a lower price

Related models

• claude-fable-5-1 — Anthropic's most capable model
• claude-opus-5 — the prior Opus release
• gpt-6-sol — OpenAI's balanced gpt-6 model