Models & pricing / gpt-6-luna · OpenAI
OpenAI

gpt-6-luna API

GPT-6 Luna is OpenAI’s efficiency-focused model for well-defined, high-volume tasks, with text and image input and text output. Evaluate current XiuAI costs and task results, and check the different tool-calling requirements of Responses and Chat Completions.[1][2]

gpt-6-luna

Model overview

Published provider specifications. Available features depend on the current API route.

API routes
OpenAI-compatible API / openai-response / openai-response-compact / Anthropic Messages / Gemini GenerateContent / openai-alpha-search

Keep high-frequency tasks well defined

OpenAI emphasizes focused tasks and efficiency. For routing content, organizing short material or extracting fields under explicit rules, fix the input format and output acceptance criteria first. Identify cases that need human or alternative-model review. High-volume positioning does not mean unlimited concurrency or remove the need for application validation.[1]

Choose the protocol for reasoning and tools together

GPT-6 Luna supports reasoning settings from none to deeper levels. OpenAI directs built-in tools and function calling with reasoning to Responses; Chat Completions supports function calling only with reasoning_effort=none. When migrating an agent, do not enable reasoning without reviewing the existing tool-message structure and protocol.[1]

Evaluate efficiency and billing separately

Use the current XiuAI pricing section to distinguish input, cache usage, output and the selected service tier, then reconcile request records. Vendor Fast, Flex, Batch, regional-processing or long-context rules are not automatically XiuAI rates. Sending a tier parameter alone does not prove the tier that served the request.[1][2]

API access

Using GPT-6 Luna through XiuAI

Use gpt-6-luna with an existing authorized API key. Check the current protocol and validate messages, reasoning and tool-result handling with a short request before increasing traffic. Image understanding can assist classification or extraction, but direct output remains text; media generation needs an available media model.[1][2]

Send a request using your existing API key.

POST /v1/chat/completions
curl 'https://router-api.xiu.ai/v1/chat/completions' \
  -H "Authorization: Bearer $XIUAI_API_KEY" \
  -H 'Content-Type: application/json' \
  --data-raw '{
  "model": "gpt-6-luna",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ]
}'
Read the API documentation

Frequently asked questions

Why can a chat tool request stop being appropriate after enabling reasoning?

OpenAI requires reasoning_effort=none for GPT-6 Luna function calling in Chat Completions. To combine reasoning and tools, use the Responses request and tool-result structure and check the current XiuAI route rather than changing only the reasoning setting.[1][2]

Is changing the ID enough to migrate from GPT-5.6 Luna?

Recheck parameters, tool protocols and representative outputs. Related efficiency positioning does not mean identical behavior or billing conditions. Keep your existing evaluation set and compare the migration using GPT-6 Luna documentation and current XiuAI pricing.[1][2]

Does vendor EU residency support mean my XiuAI request stays in the EU?

No. Vendor eligibility, processing modes and regional options belong to a specific service configuration. A model name or documentation entry does not establish the provider or region of an XiuAI request; verify the explicit integration terms when region matters.[1][2]

Related models

gpt-5.6-lunagpt-6.1-terragpt-6-sol

Sources and updates

Specifications reference provider documentation; prices and availability come from the current XiuAI catalog.

Content reviewed: 2026-11-16

[1] OpenAI GPT-6 Luna documentation[2] Current XiuAI model and pricing dataModel documentationTool callingStructured outputContext documentation