NezhaGateNezhaGate
简体中文English ✓Español한국어日本語Русский
Log in Get API Key
Chat

Claude Opus 5 (Kiro)

⧉
claude-opus-5-kiro
Model Type StandardKiro

Claude Opus 5 (Kiro) is Claude Opus 5 on the Kiro line: the same Opus 5 model, served by the Kiro line at a quarter of the price of the standard line, claude-opus-5. It is strong at complex reasoning, code engineering and agent orchestration and supports image input, tool calling and streaming; it answers directly without returning a thinking trace, reasoning_effort has no effect, PDF input is not supported and stop is ignored. Fully OpenAI-compatible: set model to claude-opus-5-kiro; the native Anthropic Messages API works too, so Claude Code can connect directly. Pay-as-you-go, failed calls never billed.

ChatCodeAgentsKiro lineNative Anthropic API
Get API Key → API Docs

Live Test · Playground

Try out Claude Opus 5 (Kiro) right here (available after login).

Input

Advanced
Web search
Memory · multi-turn
Thinking

Conversation

Start a conversationType a message below to begin

About Claude Opus 5 (Kiro)

Claude Opus 5 (Kiro) is Claude Opus 5 on the Kiro line: the same Opus 5 model, served on NezhaGate through both the OpenAI-compatible API and the native Anthropic Messages API at a quarter of the price of the standard line, claude-opus-5. It is strong at complex reasoning, code engineering and agent orchestration, with image input, tool calling and streaming. This line is provided by Kiro: it answers directly without returning a thinking trace, and reasoning_effort has no effect; when you need to see the reasoning or tune its depth, use the standard line, claude-opus-5. Set model to claude-opus-5-kiro, or point Claude Code's ANTHROPIC_BASE_URL at NezhaGate's /anthropic; pay-as-you-go, failed calls never billed.

Use cases

High-volume coding work

Code generation, refactoring and review at volume: Opus 5 capability at a quarter of the price of the standard line.

Agents and tool calling

OpenAI-format tools / tool_calls and native tool_use, including forcing a specific tool, for agents that plan in several steps and call tools repeatedly.

Claude Code, connected directly

Point ANTHROPIC_BASE_URL at NezhaGate's /anthropic, use your gateway key as x-api-key and set the model to claude-opus-5-kiro.

Image understanding

Q&A over screenshots, charts and mixed text-and-image material; the gateway converts public image links to base64 before sending them.

How to choose

Choose the Kiro line, claude-opus-5-kiro, for Opus 5 capability at a lower cost when you do not need to see the reasoning; choose the standard line, claude-opus-5, when you need a thinking trace, reasoning_effort control or PDF input; for the newer Opus, Claude Opus 5.5 is served by the same Kiro line. Switching is just the model field.

FAQ

How is the Kiro line different from the standard line?
It is the same model, Claude Opus 5. The Kiro line (claude-opus-5-kiro) costs a quarter of the standard line (claude-opus-5), but it answers directly without returning a thinking trace, reasoning_effort has no effect, PDF input is not supported and stop is ignored; the standard line keeps the thinking trace and reasoning_effort. To switch lines, change the model field.
Does it return its thinking?
No. This line answers directly, with no thinking or reasoning_content in the response; parameters such as thinking or reasoning_effort do not cause an error, but they do not produce visible reasoning either. Use claude-opus-5 when you need the thinking trace.
Which parameters work, and what are the limits?
messages, streaming, tools (tool_choice auto or a specific tool) and image input all work. temperature, top_p and stop / stop_sequences have no effect; PDF document input is not supported yet.
How do I send an image?
Put the image in the message content: image_url on the OpenAI format, an image block on the native API. Base64 and public image links both work; the gateway converts a public link to base64 before sending it.
Am I charged for failed calls?
No. Only calls that return normally are settled, at actual usage; upstream errors and timeouts never touch your balance.

Related Models

Explore other models you can integrate.

View all →
Claude Opus 5 Chat

Claude Opus 5

claude-opus-5

Next-generation flagship from Anthropic

$4.00/1M in · $20.00/1M out ↓20% View →
Claude Opus 5.5 Chat

Claude Opus 5.5

claude-opus-5-5

Anthropic's next Opus - list price 20% below Opus 5

$0.80/1M in · $4.00/1M out ↓80% View →
Claude Sonnet 5 Chat

Claude Sonnet 5

claude-sonnet-5

Anthropic's new-generation Sonnet · near-Opus coding

$1.00/1M in · $5.00/1M out ↓50% View →
Claude Fable 5 Chat

Claude Fable 5

claude-fable-5

Anthropic Fable series · narrative and long-form writing

$8.00/1M in · $40.00/1M out View →
Claude Sonnet 4.6 Chat

Claude Sonnet 4.6

claude-sonnet-4-6

Balanced, efficient chat and coding model

$1.50/1M in · $7.50/1M out ↓50% View →
GPT-6 Astra Chat

GPT-6 Astra

gpt-6-astra

Next-generation OpenAI flagship - 1.05M context

$1.00/1M in · $5.00/1M out ↓90% View →