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

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

How much does GPT-6 Astra cost on Elosia?

GPT-6 Astra is billed by usage at $10 per 1M input tokens and $50 per 1M output tokens, within your Elosia plan.

What is GPT-6 Astra's context window?

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

Is GPT-6 Astra hosted in the EU?

Yes, GPT-6 Astra is available on Elosia's EU-hosted (EdenAI) backend for enterprise organizations.

When should you choose GPT-6 Astra?

GPT-6 Astra is a flagship model, best for the most demanding reasoning, agentic and long-context tasks.

GPT-6 Astra

GPTFlagship

ThinkingTool UseVisionStructured Output

About this model

GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon...

Performance Tier

Flagship

GPT-6 Astra is a flagship model from GPT : the most capable in their lineup.

Best-in-class model from this provider. Highest performance across benchmarks, ideal for demanding tasks.

Pricing

This model is included in Elosia plans

Premium

Highest cost level. A long conversation can quickly use up your included usage or your balance.

Type	per 1M tokens
Input (prompt)	$10.00
Output (completion)	$50.00
Cache read	$1.00
Cache write	$12.50

Capabilities

Context Length1.1M

Max Output Tokens128K

TokenizerGPT

Inputfile, image, text

Outputtext

Release DateSeptember 4, 2026

Benchmarks

General Intelligence

MMLU

Not reported

GPQA Diamond

96%

Mathematics

MATH-500

Not reported

Programming

HumanEval

Not reported

SWE-bench Verified

Not reported

Reasoning

IFEval

Not reported

ARC-AGI-2

92.1%

Humanity's Last Exam

54.7%

Multimodal

MMMU-Pro

86.9%

Where does GPT-6 Astra stand?

Compare its performance index against every other model.

View the performance leaderboard

Recommended Use Cases

CodingAnalysisResearchMathematicsData Extraction

Strengths

Top model of the GPT-6 family, built for demanding end-to-end work: advanced analysis, software engineering, deep research, scientific work and document creation
Strong computer use: operates desktop environments and completes multi-step UI tasks such as forms, CRM updates and QA checks
Reliable retrieval across its full 1M-token context window, per OpenAI's long-context evaluations
Decides on its own when to ask a focused clarifying question and when to proceed on sensible assumptions
Produces polished professional artifacts (documents, slides, spreadsheets) that follow a supplied template

Limitations

OpenAI's system card reports that its chain of thought is harder to monitor than in earlier GPT models, with the model better at controlling what its reasoning reveals
Top scores are reported at maximum reasoning effort, which is slow and token-hungry; lower efforts score markedly lower
OpenAI publishes no SWE-bench or Terminal-Bench 2.x results; the ARC-AGI-2, MMMU-Pro and Humanity's Last Exam scores shown come from third-party testing
Offensive cybersecurity capabilities are gated behind a separate access program, and safeguards can pause legitimate security work