Claude Opus 5.5 is Anthropic's first model in the Claude 5.5 family, delivering frontier-level performance in agentic coding, computer use, and knowledge work. With faster output, clearer communication, and significantly improved computational efficiency, it brings advanced intelligence to demanding, multi-step tasks.

Pricing: Input 320 credits / 1M tokens (≈ $1.60), Output 1,600 credits / 1M tokens (≈ $8.00) — both 60% below official pricing. High-tier top-ups come with bonus credits that lower the effective price by about 10%.
claude-opus-5-5

Automatically append previous messages to maintain multi-turn context. May increase token usage.

Function calling
Thinking process
Max tokens
Stream Response

How can I help you today?

README

Complete guide to using claude-opus-5-5

Claude Opus 5.5 API: Stronger Coding, Smarter Agents, Less Overhead

Original image
Claude Opus 5.5 API Takes On Complex Coding Challenges
Smarter Computer Use with Opus 5.5 API
Anthropic Claude Opus 5.5 API Brings Deeper Intelligence to Knowledge Work
Lower Workload Costs with Claude Opus 5.5 API
Faster Output, Clearer Responses from Opus 5.5 API
Stronger Alignment Across Anthropic Opus 5.5 API Workflows
BenchmarkClaude Opus 5.5Claude Fable 5.1Claude Opus 5GPT-6 AstraGPT-5.6 Sol
Agentic coding — Terminal-Bench 4.0¹66.4%55.8%52.3%57.9%37.3%
Agentic coding — FrontierCode v1.1 (Main)54.4%50.3%48.0%53.3%47.5%
Agentic coding — CursorBench 4.057.8%51.8%46.6%—41.7%
Knowledge work — GDPval-AA v2.118461735170815421588
Business workflows — AutomationBench²40.0%31.4%26.9%41.4%28.8%
Multidisciplinary reasoning — Humanity’s Last Exam67.7% with tools65.6% with tools63.6% with tools57.2% with tools—
Agentic scientific research — Terminal-Bench-Science 0.1³58.7%52.6%29.0%64.6%22.4%
Computer use — OSWorld 2.081.8% partial80.7% partial74.0% partial——
Visual chart recognition — Chartography89.0% with tools88.4% with tools83.4% with tools——

Send Authorization: Bearer YOUR_API_KEY. Clients that send x-api-key, such as the Anthropic SDK, set it to Bearer YOUR_API_KEY, and anthropic-version is optional. Input and output token pricing is shown in the pricing section above.

Declare callable functions in tools, each with a name, a description and a JSON Schema input_schema. With stream set to true, tool calls arrive as tool_use blocks and input_json_delta fragments.

stream defaults to true and returns server-sent events, thinkingFlag turns thinking on, and max_tokens defaults to 4096 when left empty.

Claude Opus 5, Claude Opus 4.8 and Claude Sonnet 5.5 use the same /claude/v1/messages endpoint with their own model ids, and GET https://api.kie.ai/anthropic/v1/models lists the available ids.

Claude Opus 5.5 is the first model in Anthropic's Claude 5.5 family, designed for demanding coding, computer use, and knowledge-intensive tasks. It delivers performance comparable to Claude Fable 5.1 on most work while improving speed, computational efficiency, and communication. Developers can use Claude Opus 5.5 API to bring its supported capabilities into AI-powered applications and complex workflows.

The Anthropic Claude Opus 5.5 release was announced on September 22, 2026. As the first model in the Claude 5.5 family, it introduces improvements in agentic coding, professional knowledge work, output speed, and alignment.

Compared with Opus 5, Claude Opus 5.5 delivers stronger performance across agentic coding, computer use, and knowledge work. It also generates output more than 30% faster and costs approximately 40% less to run on typical workloads at default settings, according to Anthropic's testing. Clearer communication and improved resistance to prompt injection further distinguish the newer model.

The published Claude Opus 5.5 benchmarks demonstrate improvements across several technical and professional evaluations. The model scores 66.4% on Terminal-Bench 4.0, 54.4% on FrontierCode v1.1, and 1846 Elo on GDPval-AA v2.1. It also achieves 81.8% on the partial OSWorld 2.0 computer-use evaluation. Results depend on the benchmark configuration and evaluation settings.

Developers can explore Claude Opus 5.5 API for coding assistants, research applications, business analysis tools, and other reasoning-intensive projects. Its coding capabilities also make it relevant to creative development tasks such as browser games, JavaScript animations, interactive simulations, and 3D visualization. The specific implementation options depend on the features supported by Kie.ai's API.

Not necessarily. Kie.ai provides access to supported Claude Opus 5.5 API functionality, which may differ from the complete feature set available through Anthropic's official platform. Check the corresponding Kie.ai API documentation for available parameters, request formats, and integration options before implementing a particular workflow.

Claude Opus 5.5 API pricing on Kie.ai depends on the platform's current rates and your actual usage. Anthropic reports that Opus 5.5 reduces typical workload costs by approximately 40% compared with Opus 5 at default settings, but this figure reflects Anthropic's own evaluation rather than Kie.ai's pricing. Visit the model's pricing page on Kie.ai for the applicable rates.

You can create a Kie.ai account and check whether trial credits or other introductory offers are available for Claude Opus 5.5 API. Free access depends on the platform's current offers and applicable model pricing. Before making your first request, review the API documentation and available account balance to understand the supported functionality and usage requirements.