OpenAI starts rolling out its next-generation GPT-6 Astra model
OpenAI ships GPT-6 Astra with state-of-the-art coding, browsing, and computer use — the model that changes what agents can do.

Why it matters
A new frontier-capability model release from OpenAI with claimed SOTA across multiple domains (coding, browsing, computer use). This is the kind of model release that practitioners benchmark against and re-evaluate their stacks around; enthusiasts tracking the lab race will see this as a major inflection point.
The key facts
5 to knowGPT-6 Astra released by OpenAI
Claims state-of-the-art performance in coding, browsing, and computer use
Computer use capability (GUI interaction) represents a step toward agentic AI
Rollout beginning as of September 3, 2026
Model positioned as next-generation LLM
Go to the source
SiliconAnglesiliconangle.com
Publisher excerpt: OpenAI Group PBC today started opening access to GPT-6 Astra, its newest and most capable large language model. The company stated that the LLM demonstrates “state of the art” performance in multiple areas. The list includes coding, browsing and computer use, a term for tasks that require a model…