OpenAI's most efficient GPT-6 model for high-volume, focused tasks — vision, web search, and function calling at half the cost of its predecessor.
curl https://api.venice.ai/api/v1/chat/completions \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai-gpt-6-luna",
"messages": [{ "role": "user", "content": "Explain quantum tunneling simply." }]
}'Overview
GPT-6 Luna is OpenAI's most efficient large language model in the GPT-6 series, released on September 22, 2026. Designed for high-volume, focused tasks, it supports vision, function calling, reasoning, and web search, with a 1.05M-token context window and pricing cut in half compared to GPT-5.6 Luna.
On Venice, GPT-6 Luna runs with anonymized privacy — your prompts are never stored or used for training. This means you get the full power of OpenAI’s efficient, multimodal model without surveillance, ideal for scaling applications where cost, privacy, and input flexibility matter. You maintain sovereignty over your data while accessing one of the most cost-effective high-context models available.
Agent quickstart
The API is OpenAI-compatible: change the base URL and the model id and existing client code works unchanged.
curl https://api.venice.ai/api/v1/chat/completions \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai-gpt-6-luna",
"stream": true,
"messages": [{ "role": "user", "content": "Draft the release note." }]
}'curl https://api.venice.ai/api/v1/chat/completions \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai-gpt-6-luna",
"messages": [{ "role": "user", "content": "Find the rate limits." }],
"tools": [{
"type": "function",
"function": {
"name": "search_docs",
"description": "Search the API documentation.",
"parameters": {
"type": "object",
"properties": { "query": { "type": "string" } },
"required": ["query"]
}
}
}],
"tool_choice": "auto"
}'import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["VENICE_API_KEY"],
base_url="https://api.venice.ai/api/v1",
)
resp = client.chat.completions.create(
model="openai-gpt-6-luna",
messages=[{"role": "user", "content": "Build without permission."}],
)
print(resp.choices[0].message.content)Specifications
Assessment
Use cases
Prompting
Use 'reasoning.effort=none' for fastest, cheapest responses on simple tasks.
Include web search in your tool configuration when up-to-date results are needed.
Upload multiple images in a single request to compare or analyze them together.
Wrap structured output requirements in JSON schema to ensure consistent formatting.
Use cached input for repeated prompts — saves 90% on input token cost.
For long documents, split into chunks under 272K tokens to avoid premium pricing.
Alternatives
| Model | Best for | Context | Open weights | Price (Venice) |
|---|---|---|---|---|
| GPT-6 Luna | High-volume, cost-efficient tasks | 1M tokens | No | $0.13 in · $0.63 out / 1M |
| Claude Fable 5.1 | Balanced performance and cost | 1M tokens | No | $12 in · $60 out / 1M |
| DeepSeek V4.1 Flash | Open-weight speed | 1M tokens | Yes | $0.38 in · $1.50 out / 1M |
| Google Gemma 4 31B Instruct | Open-weight budget option | 256K tokens | Yes | $0.12 in · $0.36 out / 1M |
Choose GPT-6 Luna when you need maximum context and multimodal capabilities at the lowest possible cost, especially for high-volume, privacy-sensitive workloads. It outperforms rivals on price-to-performance for focused tasks.
Pricing
Billed per token on Venice: $0.13 per 1M input tokens and $0.63 per 1M output tokens.
New Venice accounts include a free daily allowance and 500 welcome credits — no credit card required.
Getting a key
Create a key in API settings. Nothing else is required to start.
Export it as VENICE_API_KEY so the snippets above run unedited.
Point an existing OpenAI client at https://api.venice.ai/api/v1. The scheme is part of the value: an OpenAI client given a bare host does not resolve it.
Pass openai-gpt-6-luna as the model and send the request.
FAQ
GPT-6 Luna is OpenAI's most efficient model in the GPT-6 lineup, released on September 22, 2026. It's designed for high-volume, focused tasks with a 1.05M-token context, vision, web search, and function calling, all at half the price of GPT-5.6 Luna.
On Venice, GPT-6 Luna costs $0.13 per 1M input tokens and $0.63 per 1M output tokens. Cached input is just $0.01 per 1M tokens, making repeated queries extremely cost-effective.
No. GPT-6 Luna is a proprietary model owned by OpenAI. It is not open source or open weights, so it cannot be self-hosted or fine-tuned. However, you can use it via API on Venice without a subscription.
Yes. GPT-6 Luna supports multiple image inputs in JPEG, PNG, GIF, and WebP formats, with a maximum size of 20MB per image. This enables document analysis, visual reasoning, and multimodal workflows.
GPT-6 Luna has a 1,050,000-token context window, allowing it to process extremely long documents, codebases, or conversations in a single request.
GPT-6 Sol is more powerful and better for complex coding and agentic workflows, while GPT-6 Luna is optimized for cost efficiency and high-volume, focused tasks. Sol costs more ($2/$10 per 1M tokens) and scores higher on benchmarks; Luna is cheaper ($0.13/$0.63) and slightly regressed in coding tasks but ideal for scaling.
Yes. GPT-6 Luna supports function calling, web search, and structured output, making it suitable for building agents that interact with external tools and APIs.
One key, free to start, no credit card.