Claude Opus 5.5 API
anthropic/claude-opus-5.5Claude Opus 5.5 turns text and images into code and analysis, combining a million-token context with adaptive reasoning for multi-step work. Provide source files, charts, and acceptance criteria to keep its responses focused on your objective and key constraints.
Claude Opus 5.5
Anthropic · chat-completions
Quick start
Send a POST request to /v1/chat/completions with anthropic/claude-opus-5.5. Choose JSON or SSE with stream, and receive generated content and usage.
- Endpoint
- POST /v1/chat/completions
- Model ID
- anthropic/claude-opus-5.5
- Protocol
- Chat Completions
Request fields
| Field | Requirement | Description |
|---|---|---|
| model | Required | anthropic/claude-opus-5.5 |
| messages | Required | A non-empty array with role and content in each message; supply system instructions in a system-role message. |
| max_tokens | Optional | A positive integer controlling the response budget, including thinking tokens. Model maximum: 128,000 tokens; playground: 1–65,536, default 4,096. |
| max_completion_tokens | Optional | An output token budget field for Chat Completions; choose one output-budget field per request. |
| temperature | Optional | Sampling temperature; the playground offers 0–2 and defaults to 1. |
| top_p | Optional | Nucleus sampling probability; the playground offers 0–1 and defaults to 1. |
| stream | Optional | Defaults to false for complete JSON; set true for an SSE event stream. |
Response and usage
Successful responses include generated content and usage. Billing settles from actual input, output, and cache tokens.
Streaming
Set stream: true to read SSE. Finalize usage accounting from the terminal usage event.
curl --request POST \
--url "https://api.vidgo.ai/v1/chat/completions" \
--header "Authorization: Bearer $VIDGO_API_KEY" \
--header 'Content-Type: application/json' \
--data '{
"model": "anthropic/claude-opus-5.5",
"messages": [
{
"role": "system",
"content": "You are a helpful assistant."
},
{
"role": "user",
"content": "Explain why deterministic retries matter in distributed systems."
}
],
"max_tokens": 4096,
"temperature": 1,
"top_p": 1,
"stream": false
}'Store API keys on your server and call the model from server-side code.
Continue with
Related Models
Claude Opus 5.5 API Frequently Asked Questions
What is the Claude Opus 5.5 API?
Claude Opus 5.5 is an Anthropic model that turns text and images into code and analysis for long-running coding and knowledge work. It combines a million-token context, adaptive reasoning, and visual understanding for cross-file changes, long documents, and charts. Supply task constraints and adjust reasoning depth with output_config.effort in Claude Messages. You can call it programmatically or try text conversations in the playground above.
How does Claude Opus 5.5 help with code migration?
Give Claude Opus 5.5 the relevant source files, target architecture, compatibility requirements, and test results. Ask it to trace dependencies and outline migration steps before proposing changes and verification. Its long context holds connected task evidence for reviewing multi-file changes across subsequent turns.
How large is the Claude Opus 5.5 context window?
Claude Opus 5.5 has a 1,000,000-token context window for source files, documents, and conversation history. Keep file names, section names, and citation locations in your material so its analysis can refer back to the evidence from different sources.
How does Claude Opus 5.5 analyze charts and screenshots?
Combine text and image blocks in the content of a Claude Messages request. Specify the values to read, regions to compare, or interface state to explain. Claude Opus 5.5 uses the visual content and your question to produce text analysis; see the API view for the request format.
How do I adjust Claude Opus 5.5 reasoning effort?
Set output_config.effort in a Claude Messages request to low, medium, high, xhigh, or max; medium is the default. Adaptive thinking is always active in Claude Opus 5.5. Choose effort for the complexity of your task and leave room in max_tokens for both thinking and the answer.
How do I set the Claude Opus 5.5 output budget?
The Claude Opus 5.5 model maximum is 128,000 output tokens. This playground offers Max Tokens from 1 to 65,536, with a default of 4,096. The budget includes thinking tokens, so choose a value for the task complexity and expected answer length, then inspect usage for actual consumption.
How does Claude Opus 5.5 reuse prompt caching?
Set cache_control on cacheable content in a Claude Messages request with type: ephemeral and ttl: 5m or 1h, and keep the repeated content prefix stable. Claude Opus 5.5 has a minimum cacheable prompt length of 512 tokens. Cache reads and the two cache-write tiers are metered separately at the rates in this page’s pricing table.