Claude Opus 5.5 (AWS) API
https://api.hiapi.ai /v1/chat/completions This model uses an OpenAI-compatible Chat Completions endpoint. The same HiAPI API key can call enabled models in your account group; image, video, and audio models use /v1/tasks and a different request shape.
Model overview
| Model name | Claude Opus 5.5 (AWS) |
|---|---|
| Provider | Anthropic |
| Context / max output | 1M / 128K tokens |
| Endpoint | POST /v1/chat/completions |
| Thinking | Always on · default medium effort |
| Pricing | Live HiAPI token rates |
Anthropic's Claude Opus 5.5 served through HiAPI's official AWS route for long-running agentic coding and knowledge work. This route offers higher stability for production workloads, with always-on adaptive thinking controlled by effort.
Production guidance
- Use model=claude-opus-5-5@aws with POST /v1/chat/completions. The AWS route and the standard claude-opus-5-5 route are billed independently; pick one model ID per workload.
- Do not send thinking.type=disabled or a manual thinking budget. Omit thinking and use output_config.effort.
- Allowed effort values are low, medium, high, xhigh, and max. Medium is the provider default.
- Use tool_choice=auto. Forced any and named-tool selection return an upstream error.
Best suited for
Complex coding, knowledge work, and multi-step tool loops where effort can trade cost and depth.
messagesoutput_config.efforttoolsRequest parameters
model string required Use this exact public model ID.
messages array required Ordered system, user, assistant, and tool messages. Preserve returned reasoning_details unchanged in tool continuations.
stream boolean optional Set true for Server-Sent Events.
output_config.effort enum optional The only public thinking-depth control. Thinking is always on.
max_tokens integer optional Hard limit shared by thinking and visible output.
system content cache_control object optional Optional five-minute prompt-cache write. One-hour cache writes are billed per the pricing page when the upstream reports them.
tools array optional OpenAI-compatible function definitions.
tool_choice enum optional Forced any or named-tool choices are rejected by this model.
API examples
Request examples
Read content and reasoning deltas until [DONE].
{
"model": "claude-opus-5-5@aws",
"messages": [
{
"role": "user",
"content": "Analyze the trade-offs between a monolith and microservices."
}
],
"stream": true,
"output_config": {
"effort": "medium"
},
"max_tokens": 4096
}Let the model choose a function with tool_choice=auto, then replay the tool result with the original reasoning metadata.
{
"model": "claude-opus-5-5@aws",
"messages": [
{
"role": "user",
"content": "Check task demo-123."
}
],
"stream": false,
"output_config": {
"effort": "low"
},
"max_tokens": 4096,
"tools": [
{
"type": "function",
"function": {
"name": "get_task_status",
"description": "Look up a task by ID.",
"parameters": {
"type": "object",
"properties": {
"task_id": {
"type": "string"
}
},
"required": [
"task_id"
],
"additionalProperties": false
}
}
}
],
"tool_choice": "auto"
}Add cache_control to a stable system text block and verify writes or hits in usage.
{
"model": "claude-opus-5-5@aws",
"messages": [
{
"role": "system",
"content": [
{
"type": "text",
"text": "Answer as a concise architecture reviewer.",
"cache_control": {
"type": "ephemeral",
"ttl": "5m"
}
}
]
},
{
"role": "user",
"content": "Review this migration plan."
}
],
"stream": false,
"output_config": {
"effort": "medium"
},
"max_tokens": 4096
}Response schema
Read visible text from choices[0].message.content and readable thinking from reasoning_content when returned. Preserve reasoning_details unchanged for tool continuations. Usage is the billing source of truth.
{
"id": "chatcmpl_example",
"object": "chat.completion",
"model": "claude-opus-5-5@aws",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "Start with a migration boundary that can be rolled back.",
"reasoning_content": "First compare deployment and data risks."
},
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 32,
"completion_tokens": 48,
"total_tokens": 80
}
}- Read choices[0].message.content for visible text.
- For streams, accumulate content and reasoning deltas until [DONE].
- Read input, output, and cache usage; thinking counts as output.
FAQ
What is Claude Opus 5.5 (AWS)?
It is the same Anthropic Opus 5.5 model served through the HiAPI official AWS route, with a documented 1M context window and 128K maximum output. The AWS route is positioned for higher stability in production.
Which endpoint and model ID should I use?
Use POST /v1/chat/completions with model=claude-opus-5-5@aws.
How is it different from Claude Opus 5.5?
Same model and request contract on the official AWS upstream line, with independent capacity and live rates. Do not switch model IDs mid-conversation.
Can I disable thinking?
No. Thinking is always on. Requests that send disabled or a manual budget are rejected; use output_config.effort instead.
Which effort values are supported?
Low, medium, high, xhigh, and max are supported. Medium is the default.
How are tokens billed?
Input, output, five-minute cache writes, one-hour cache writes when reported, and cache reads use the live rates on the model and pricing pages. Thinking tokens count as output. View live pricing.