Is GPT-6 Luna available with Zero Data Retention on Elosia?

GPT-6 Luna runs on Elosia's standard backend; ZDR is available on many other models in the catalog.

How much does GPT-6 Luna cost on Elosia?

GPT-6 Luna is billed by usage at $0.09999999999999999 per 1M input tokens and $0.5 per 1M output tokens, within your Elosia plan.

What is GPT-6 Luna's context window?

GPT-6 Luna supports a context window of up to 1,050,000 tokens.

Is GPT-6 Luna hosted in the EU?

GPT-6 Luna is not available on Elosia's EU-hosted (EdenAI) backend; it runs on the default backend.

When should you choose GPT-6 Luna?

GPT-6 Luna is a compact model, best for high-throughput, latency-sensitive or cost-sensitive workloads.

GPT-6 Luna

GPTCompact

ThinkingTool UseVisionStructured Output

About this model

GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and lightweight agentic...

Performance Tier

Compact

GPT-6 Luna is a compact model from GPT : optimized for speed and affordability.

Small, fast, and affordable. Optimized for speed and low cost, great for high-volume or simple tasks.

Pricing

This model is included in Elosia plans

Affordable

Low cost. Suitable for sustained use and high-volume interactions.

Type	per 1M tokens
Input (prompt)	$0.100
Output (completion)	$0.500
Cache read	$0.010
Cache write	$0.125

Capabilities

Context Length1.1M

Max Output Tokens128K

TokenizerGPT

Inputfile, image, text

Outputtext

Release DateSeptember 22, 2026

Benchmarks

General Intelligence

MMLU

Not reported

Mathematics

MATH-500

Not reported

Programming

HumanEval

Not reported

Reasoning

ARC-AGI-2

31.4%

Humanity's Last Exam

38.5%

Multimodal

MMMU-Pro

75.5%

Where does GPT-6 Luna stand?

Compare its performance index against every other model.

View the performance leaderboard

Recommended Use Cases

General ChatData ExtractionSummarizationCustomer SupportCoding

Strengths

Fast, cost-efficient tier of the GPT-6 family, built for high-volume and latency-sensitive workloads such as chat, classification and lightweight agents
1M-token context window, uncommon for a model of this class
Solid agentic coding and computer use for its class when run at high reasoning effort
Multimodal (text, image, file) input with native tool calling and structured outputs

Limitations

Abstract reasoning drops sharply at lower reasoning effort
Third-party testing measures a high hallucination rate on questions it cannot answer: it guesses more often than it abstains
OpenAI publishes no standard academic benchmarks for it; the scores shown come from third-party testing