OpenAI Advanced Available in FlowHunt

GPT-6 Luna

GPT-6 Luna is an Advanced-tier AI model developed by OpenAI and released in September 2026. It is a proprietary closed-weight model. The 1.05M-token context window is large enough to hold entire codebases, lengthy technical documents, or extended agentic sessions without truncation.

Read more
OpenAI introduced Luna on September 22, 2026 alongside GPT-6 Sol as part of the GPT-6 family, and describes it as its fastest and most cost-efficient model yet. It is the small tier of the family, below GPT-6 Astra and Sol. OpenAI treats it as High capability in cybersecurity and in biological and chemical domains.
Try in FlowHunt Back to AI Leaderboard

Benchmarks

Each score appears once, with its rank among tracked models and who measured it.

Compared with Advanced peers on the same scale
  • GPT-6 Lunathis model
  • Mistral Large 4this model leads on 1 of 3
  • Claude Sonnet 5.5this model leads on 0 of 3
Verified independent evaluator, counted in the ranking
Overall

AA Intelligence Index

Independent composite of agent, coding, knowledge and science tests by Artificial Analysis (v4).

Source ↗

GPT-6 Luna 38.1 Mistral Large 4 38.4 Claude Sonnet 5.5 56.0
#24of 45 Verified
Agents

Terminal-Bench 4.0

Multi-step professional tasks done by an agent in a command-line terminal.

Source ↗

GPT-6 Luna 12.6% Mistral Large 4 26.8% Claude Sonnet 5.5 63.6%
#22of 36 Verified
Reasoning

Humanity's Last Exam

Expert-written questions across 100+ subjects, no tools.

Source ↗

GPT-6 Luna 38.5% Mistral Large 4 35.0% Claude Sonnet 5.5 55.0%
#28of 44 Verified

ARC-AGI-2

Novel visual puzzles that cannot be memorised.

Source ↗

GPT-6 Luna 59.3% Mistral Large 4 No score Claude Sonnet 5.5 No score
#21of 27 Verified

What to Use GPT-6 Luna For

Fast, High-Volume Tasks

OpenAI positions Luna as its fastest and most cost-efficient model, which suits high-volume, latency-sensitive work.

Agent Sub-Tasks

Use it for the cheaper steps inside a larger agent workflow, and keep a larger model for the hardest reasoning.

Long-Document Processing

With a 1.05M-token context window, GPT-6 Luna can read long documents and large codebases in a single prompt.

Strengths and limitations

Strengths

  • Fast and cost-efficient, per OpenAI
  • Long context (1.05M)

Limitations

  • Closed-weight, so it cannot be self-hosted or fine-tuned on private data
  • Small tier of the GPT-6 family, below Astra and Sol

Use it in FlowHunt

Use GPT-6 Luna in any flow, next to your own tools and knowledge sources.

  1. 1Open a flow, or create one from a template
  2. 2Add an AI Agent or Generator component
  3. 3Pick GPT-6 Luna in the model selector

Related AI Models

About OpenAI San Francisco, CA · Founded 2015 · openai.com

OpenAI was founded in December 2015 as a nonprofit AI research laboratory by Sam Altman, Elon Musk, Greg Brockman, Ilya Sutskever, John Schulman, and Wojciech Zaremba, with the mission of developing artificial general intelligence (AGI) that benefits all of humanity. In 2019 it created a capped-profit subsidiary, OpenAI LP, to raise institutional capital and scale model training. The company is known for the GPT series of language models (GPT-1 through GPT-6.1), the DALL-E image generation series, Whisper speech recognition, and Sora video generation. OpenAI operates the ChatGPT consumer application, a developer API, and enterprise services through Microsoft Azure. As of 2026 it maintains a close strategic partnership with Microsoft, which is the company's largest investor and cloud provider.

openai.com ↗

Browse the Full AI Model Leaderboard

Compare every tracked model on coding, agent and reasoning benchmarks, with each score marked verified or self-reported.

← Back to AI Leaderboard Try in FlowHunt

Frequently asked questions