OpenAI Launches GPT-6 Astra API with 1M Token Context and New Pricing Tiers

·4 views

OpenAI released GPT-6 Astra on September 3, 2026, initially to a limited set of enterprise customers before rolling it out to ChatGPT Plus, Pro, Business, and Enterprise users, as well as via API on Azure and AWS Bedrock. The model, identified as gpt-6-astra, supports a context window of 1,050,000 tokens and a maximum output of 128,000 tokens, with a knowledge cutoff of April 30, 2026. Standard API pricing is set at $10 per million input tokens and $50 per million output tokens, making it roughly 2.5 times more expensive than GPT-5.6 Sol, which remains available at promotional rates until at least November 21, 2026. OpenAI has removed several legacy parameters — including temperature, top_p, and logprobs — and replaced them with structured reasoning effort levels ranging from low to max. The model is also the first OpenAI has classified as critical for cybersecurity capability, a designation that affects its default behavior through the API.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Developer Argues Rejecting AI Tools Is Pride, Not Principle

A software developer has published a rebuttal to a widely read article in which programmer Brett Codes announced he was done using AI tools. The author argues that developers are paid to deliver value, not merely to write code by hand, and that refusing AI assistance is driven by false pride rather than rational reasoning. Drawing comparisons to past technologies like frameworks, containers, and modern programming languages, the writer contends that AI is simply the latest productivity abstraction in a long line of tools developers have adopted. While acknowledging real flaws in current AI tooling and misuse by non-technical users, the author insists these are problems to be addressed and improved upon, not reasons for outright rejection. The piece concludes that an unwillingness to adapt runs counter to the spirit of technological progress that has defined software development for decades.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Why Problem-Solving, Not Coding, Is the Key to Career Safety in the AI Era

Software engineers are increasingly anxious about AI making their skills obsolete, even as AI coding tools grow rapidly more capable. Industry voices like John Carmack and Linus Torvalds argue that coding itself was never the core value — problem-solving and technical judgment are. As the cost of producing code continues to fall, businesses will care less about how code is written and more about the outcomes it delivers. Engineers who anchor their identity to writing code risk being left behind, while those who focus on portable, outcome-driven skills stand to remain relevant. The key to career safety, the article argues, lies not in being irreplaceable but in continuously shifting one's value toward higher-order thinking.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Vibe Coding Builds Apps Fast, But Turning Them Into Revenue Requires Different Skills

AI coding tools now allow developers to build functional apps with authentication, payments, and deployment in a single weekend, but speed of development does not translate automatically into business success. A Stack Overflow 2025 survey found 84% of developers use or plan to use AI tools, yet the core challenges of identifying real problems, reaching customers, and earning trust remain unchanged. Experts argue developers should shift their thinking from 'what should I build' to 'what problem are people already paying to solve,' targeting workflows where businesses spend money on manual or repetitive tasks. Research methods such as scanning Reddit, GitHub Issues, job postings, and freelancer marketplaces can surface genuine pain points before a single line of code is written. The broader message is that vibe coding lowers the technical barrier to entry, but sustainable income still depends on traditional business fundamentals like validated demand and clear value proposition.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

How Enterprises Use Runtime Control Planes to Govern Autonomous AI Agents

Autonomous AI agents pose unique security risks that traditional API controls cannot address, as they independently select tools, execute multi-step plans, and interact with enterprise systems in non-deterministic ways. Unlike passive chatbots, agents can read, write, and execute commands across databases and cloud infrastructure, creating vulnerabilities such as excessive privilege delegation and prompt injection attacks. To counter these risks, enterprise teams are adopting centralized AI gateways that enforce access limits, model routing, and spend controls at runtime rather than embedding governance logic inside agent code. Bifrost, an open-source AI gateway built in Go by Maxim AI, enforces policies across more than 1,000 models while adding only 11 microseconds of latency at 5,000 requests per second. Governance frameworks also extend to Model Context Protocol settings, restricting which tools and file systems an agent can invoke to prevent unauthorized operations triggered by malicious external content.

0 comments Read more at DEV Community