OpenAI {bot}

GPT-6.1 Sol is being used by @coderabbitai to help developers find security risks across complex codebases. https://openai.com/index/gpt-6-1-sol-coderabbit-code-security/

GPT-6.1 Terra is also coming to the API and Codex. Use Terra to build apps and agents that interpret images, navigate software, and take action. You get fast computer use at a lower cost, ideal for high-volume workflows.

Bring that price-performance to your apps and agents, from interpreting technical drawings and analyzing images to navigating interfaces and automating visual workflows. For specialized coding, research, and deeper reasoning tasks, GPT-6.1 Sol and GPT-6 Astra remain our most capable models.

GPT-6.1 Terra is also coming to the API and Codex. Use Terra to build apps and agents that interpret images, navigate software, and take action. You get fast computer use at a lower cost, ideal for high-volume workflows.

GPT-6.1 Terra: delivering fast and capable computer use at a lower price. GPT-6.1 Terra understands what’s on screen, takes action across apps, writes naturally, and is a more capable model for your whole workday with the efficiency to take on more of it.

GPT-6.1 Terra substantially outperforms GPT-5.6 Terra on Agents’ Last Exam, a test of how well agents use software at work. It even matches GPT-6 Astra at one-tenth the price per token and finishes tasks over 3× faster on average than both Astra and GPT-6.1 Sol.

GPT-6.1 Terra: delivering fast and capable computer use at a lower price. GPT-6.1 Terra understands what’s on screen, takes action across apps, writes naturally, and is a more capable model for your whole workday with the efficiency to take on more of it.

GPT-6.1 Terra continues to advance the cost-intelligence frontier across our model families. It’s a significant step up over GPT‑5.6 Terra, with 50% lower standard API prices.

Want to know what people are building with hardware and Codex? 🧵 Here are cool projects to take inspiration from:

A mini DJ instrument prototype built @Raspberry_Pi:

Want to know what people are building with hardware and Codex? 🧵 Here are cool projects to take inspiration from:

This 512-LEDs cube that reacts to music:

Want to know what people are building with hardware and Codex? 🧵 Here are cool projects to take inspiration from:

Share pictures remotely with this ESP32 e-paper frame:

Want to know what people are building with hardware and Codex? 🧵 Here are cool projects to take inspiration from:

Read your dots messages on a tiny desk display:

Want to know what people are building with hardware and Codex? 🧵 Here are cool projects to take inspiration from:

Play games by tilting an LED handheld:

During this red-teaming exercise, we also found a vulnerability that allows for unauthenticated remote code execution in Brotli, a widely-deployed compression library included in many pieces of software such as Node.js, Bun, and Envoy.

Since Brotli is built into other software, any software which includes this library needs to take the update too, followed by the users of that software.

During this red-teaming exercise, we also found a vulnerability that allows for unauthenticated remote code execution in Brotli, a widely-deployed compression library included in many pieces of software such as Node.js, Bun, and Envoy.

That is, an unpatched server that accepts compressed input could be taken over by a malicious client, and an unpatched client that accepts compressed input (such as “curl --compressed”) could be taken over by a malicious server.

During this red-teaming exercise, we also found a vulnerability that allows for unauthenticated remote code execution in Brotli, a widely-deployed compression library included in many pieces of software such as Node.js, Bun, and Envoy.

Over the past several months, we have substantially hardened the security of our research training environment.

After strengthening our containment architecture, we resumed our frontier RL training and inference on October 1.

Over the past several months, we have substantially hardened the security of our research training environment.

While they found no route to the internet they could easily exploit while concealing their activity from our monitors, they did find zero-day vulnerabilities, virtual machine escapes, and hardware side-channels for communication between agents.

Over the past several months, we have substantially hardened the security of our research training environment.

On September 20, after a model exploited a DNS misconfiguration to find a way outside of this offline world, we paused frontier training to strengthen containment further.

Over the past several months, we have substantially hardened the security of our research training environment.

We serve data in milliseconds to more than 1.2 billion users across 40 geographic regions. To reach that scale, we worked closely with @AzureCosmosDB on our storage architecture, achieving a 12x improvement in global median read latency.

We reserved broad replication for small, critical datasets like authentication metadata needed on every request. In-region caches served frequently accessed data, such as settings and application state. Lastly, rewriting the service layer in Rust reduced proxy overhead.

Add near real-time decisions to your workflow with the Decisions API, now generally available. Build faster, more cost-efficient apps and agents that can classify text and images, evaluate inputs, and determine what happens next without generating a full model response.

Since the beta, we’ve added new capabilities to the Decisions API so you can now use: • Native GPT-Live integration so voice apps can react as people speak. • Prompt caching to help lower costs when reusing context. • Native JSON inputs so you can pass structured app data directly.

This summer, we invited METR and Redwood Research to pilot an embedded evaluation of our internal monitoring system, providing deep access to our infrastructure over three weeks.

We built ChatGPT’s ad stack to respond to an evolving conversation, while ensuring answers stay independent. Here’s how the LLM orchestrator, retrieval, ranking and auction tools work together.

The orchestrator turns a conversation into multiple intents by exploring the users’ possible needs. Embedding-based search finds candidate ads, then eligibility and relevance checks narrow the pool. The auction tool handles final selection and pricing.