GPT-6 Luna targets chat, classification, and lightweight agent workflows. Higher reasoning effort also supports more demanding coding and computer-use tasks. Released on September 22, 2026, it provides a 1,050,000-token context window and up to 128,000 output tokens. It accepts text, images, and files and produces text, with reasoning, tool calling, structured outputs, and prompt caching.
1.1M context window
128K max output tokens
Released: 2026-09-22
Protocols:OpenAIAnthropic
Capabilities:VisionFunction CallingReasoningPrompt CachingPDF InputWeb Search
Capabilities
- Long context1,050,000-token context window and up to 128,000 output tokens.
- Multimodal inputText, images, and files such as PDFs as input; text as output.
- Reasoning and toolsAdjustable reasoning effort, function calling, and JSON Schema structured outputs.
- Prompt cachingSeparate usage-based rates for cache reads and cache writes.
Pricing
PriceRate 0.3x
HaoAI
Official
Input
$0.03/M
$0.1/M
Output
$0.15/M
$0.5/M
Cache read
$0.003/M
$0.01/M
Cache write
$0.037/M
$0.125/M
Web search
$0.003/query
$0.01/query
Code Examples
from openai import OpenAI
client = OpenAI(
base_url="https://api.hao.ai/v1",
api_key="YOUR_HAOAI_API_KEY",
)
response = client.chat.completions.create(
model="openai/gpt-6-luna",
messages=[
{"role": "user", "content": "Hello!"}
],
)
print(response.choices[0].message.content)Related Models
Frequently asked questions
HaoAI bills based on actual request usage. Input, output, cache, and any other applicable rates follow the live pricing table on this page. Final request costs are available in the console.