OpenAI API Prompt Caching: How Shared Prefixes Affect Cost and Latency

OpenAI API prompt caching is a feature for requests that begin with the same prompt prefix. When a model processes input, it calculates intermediate key-value (KV) states that help it refer to earlier tokens as it processes more

OpenAI is retiring ChatGPT Classic app, will soon combine Chat and Work tabs in new app

ChatGPT Classic is approaching retirement as OpenAI tests a unified Chat and Work experience and prepares shared ChatGPT and Codex usage limits.

ChatGPT Release Notes: What changed in October 2026

A source-based review of the October 2026 ChatGPT updates, covering Slides, GPT-6, voice models, audio uploads, Codex predictions, and availability by plan.

OpenAI releases GPT-6.1 Sol at a fifth of GPT-6 Astra’s token prices

GPT-6.1 Sol costs significantly less than GPT-6 Astra, but developers should estimate the cost of their specific workload before choosing between them.

Latest ChatGPT guides

Practical guides and source-checked AI updates

Coding Sessions Are Longer and Use More Context. Claude Opus 5.5 Is Built with That in Mind.

This source-based guide explains Coding sessions are longer and use more context. Claude Opus 5.5 is built with that in mind., what the official first-party material supports, and what readers should verify before relying on it.

Introducing Claude Opus 4.7

Anthropic’s latest Opus model is designed for difficult coding tasks, sustained reasoning, higher-resolution vision, stronger output verification, and new cybersecurity safeguards.

OpenAI’s new ChatGPT personal finance experience brings accounts, spending, and goals together

OpenAI has added a personal finance feature to ChatGPT. It can connect financial accounts, organize spending data, remember financial goals, and answer questions based on the user's context. Despite the Wall Street framing, the announcement describes a consumer feature, not a product designed specifically for bankers.

OpenAI API Cost Optimization: A focused guide to cost and throughput choices

Learn how model selection, prompt caching, Batch, Flex processing, token counting, and related OpenAI API cost considerations affect your analysis without assuming undocumented savings.

Mermaid Diagrams in ChatGPT for iOS Codex Tasks: What Is Supported

OpenAI says Mermaid diagrams render inline in ChatGPT for iOS Codex task transcripts. This guide explains the confirmed scope, practical uses, and limits.

OpenAI’s new ChatGPT personal finance experience brings accounts, spending, and goals together

OpenAI has added a personal finance feature to ChatGPT. It can connect financial accounts, organize spending data, remember financial goals, and answer questions based on the user's context. Despite the Wall Street framing, the announcement describes a consumer feature, not a product designed specifically for bankers.

ChatGPT Sites Guide: Build, Review, and Publish a Hosted Site

A source-grounded guide to ChatGPT Sites, covering suitable projects, version review, deployment, sharing controls, storage, limits, updates, and security.

GPT-5.6 Sol vs Terra vs Luna: Which Model to Use

An official-source comparison of GPT-5.6 Sol, Terra, and Luna, including July 30 pricing, availability, safety context, and a practical task-selection method.

Popular guides

OpenAI API Prompt Caching: How Shared Prefixes Affect Cost and Latency

OpenAI API prompt caching is a feature for requests that begin with the same prompt prefix. When a model processes input, it calculates intermediate key-value (KV) states that help it refer to earlier tokens as it processes more

OpenAI is retiring ChatGPT Classic app, will soon combine Chat and Work tabs in new app

ChatGPT Classic is approaching retirement as OpenAI tests a unified Chat and Work experience and prepares shared ChatGPT and Codex usage limits.

ChatGPT Release Notes: What changed in October 2026

A source-based review of the October 2026 ChatGPT updates, covering Slides, GPT-6, voice models, audio uploads, Codex predictions, and availability by plan.

OpenAI releases GPT-6.1 Sol at a fifth of GPT-6 Astra’s token prices

GPT-6.1 Sol costs significantly less than GPT-6 Astra, but developers should estimate the cost of their specific workload before choosing between them.

Coding Sessions Are Longer and Use More Context. Claude Opus 5.5 Is Built with That in Mind.

This source-based guide explains Coding sessions are longer and use more context. Claude Opus 5.5 is built with that in mind., what the official first-party material supports, and what readers should verify before relying on it.

Follow PChatGPT

Get new guides by email

ChatGPT Updates

OpenAI News

Subscribe

ChatGPT
AI Tools

OpenAI API Prompt Caching: How Shared Prefixes Affect Cost and Latency

OpenAI API prompt caching is a feature for requests that begin with the same prompt prefix. When a model processes input, it calculates intermediate key-value (KV) states that help it refer to earlier tokens as it processes more

OpenAI is retiring ChatGPT Classic app, will soon combine Chat and Work tabs in new app

ChatGPT Classic is approaching retirement as OpenAI tests a unified Chat and Work experience and prepares shared ChatGPT and Codex usage limits.

ChatGPT Release Notes: What changed in October 2026

A source-based review of the October 2026 ChatGPT updates, covering Slides, GPT-6, voice models, audio uploads, Codex predictions, and availability by plan.

OpenAI releases GPT-6.1 Sol at a fifth of GPT-6 Astra’s token prices

GPT-6.1 Sol costs significantly less than GPT-6 Astra, but developers should estimate the cost of their specific workload before choosing between them.

Coding Sessions Are Longer and Use More Context. Claude Opus 5.5 Is Built with That in Mind.

This source-based guide explains Coding sessions are longer and use more context. Claude Opus 5.5 is built with that in mind., what the official first-party material supports, and what readers should verify before relying on it.

Introducing Claude Opus 4.7

Anthropic’s latest Opus model is designed for difficult coding tasks, sustained reasoning, higher-resolution vision, stronger output verification, and new cybersecurity safeguards.

AI Workflows

OpenAI API Prompt Caching: How Shared Prefixes Affect Cost and Latency

OpenAI API prompt caching is a feature for requests that begin with the same prompt prefix. When a model processes input, it calculates intermediate key-value (KV) states that help it refer to earlier tokens as it processes more

OpenAI is retiring ChatGPT Classic app, will soon combine Chat and Work tabs in new app

ChatGPT Classic is approaching retirement as OpenAI tests a unified Chat and Work experience and prepares shared ChatGPT and Codex usage limits.

ChatGPT Release Notes: What changed in October 2026

A source-based review of the October 2026 ChatGPT updates, covering Slides, GPT-6, voice models, audio uploads, Codex predictions, and availability by plan.

OpenAI releases GPT-6.1 Sol at a fifth of GPT-6 Astra’s token prices

GPT-6.1 Sol costs significantly less than GPT-6 Astra, but developers should estimate the cost of their specific workload before choosing between them.

Practical Guides

Recent guides
Latest

OpenAI API Prompt Caching: How Shared Prefixes Affect Cost and Latency

OpenAI API prompt caching is a feature for requests that begin with the same prompt prefix. When a model processes input, it calculates intermediate key-value (KV) states that help it refer to earlier tokens as it processes more

OpenAI is retiring ChatGPT Classic app, will soon combine Chat and Work tabs in new app

ChatGPT Classic is approaching retirement as OpenAI tests a unified Chat and Work experience and prepares shared ChatGPT and Codex usage limits.

ChatGPT Release Notes: What changed in October 2026

A source-based review of the October 2026 ChatGPT updates, covering Slides, GPT-6, voice models, audio uploads, Codex predictions, and availability by plan.

OpenAI releases GPT-6.1 Sol at a fifth of GPT-6 Astra’s token prices

GPT-6.1 Sol costs significantly less than GPT-6 Astra, but developers should estimate the cost of their specific workload before choosing between them.

Coding Sessions Are Longer and Use More Context. Claude Opus 5.5 Is Built with That in Mind.

This source-based guide explains Coding sessions are longer and use more context. Claude Opus 5.5 is built with that in mind., what the official first-party material supports, and what readers should verify before relying on it.

Introducing Claude Opus 4.7

Anthropic’s latest Opus model is designed for difficult coding tasks, sustained reasoning, higher-resolution vision, stronger output verification, and new cybersecurity safeguards.

OpenAI’s new ChatGPT personal finance experience brings accounts, spending, and goals together

OpenAI has added a personal finance feature to ChatGPT. It can connect financial accounts, organize spending data, remember financial goals, and answer questions based on the user's context. Despite the Wall Street framing, the announcement describes a consumer feature, not a product designed specifically for bankers.

OpenAI API Cost Optimization: A focused guide to cost and throughput choices

Learn how model selection, prompt caching, Batch, Flex processing, token counting, and related OpenAI API cost considerations affect your analysis without assuming undocumented savings.

Mermaid Diagrams in ChatGPT for iOS Codex Tasks: What Is Supported

OpenAI says Mermaid diagrams render inline in ChatGPT for iOS Codex task transcripts. This guide explains the confirmed scope, practical uses, and limits.

OpenAI Codex for Knowledge Work: Research, Documents, and No-Code Workflows

A source-grounded guide to OpenAI Codex for knowledge work, including research, documents, data analysis, plugins, annotations, Sites, examples, and limits.

AI Productivity

OpenAI is retiring ChatGPT Classic app, will soon combine Chat and Work tabs in new app

ChatGPT Classic is approaching retirement as OpenAI tests a unified Chat and Work experience and prepares shared ChatGPT and Codex usage limits.

ChatGPT Release Notes: What changed in October 2026

A source-based review of the October 2026 ChatGPT updates, covering Slides, GPT-6, voice models, audio uploads, Codex predictions, and availability by plan.

OpenAI releases GPT-6.1 Sol at a fifth of GPT-6 Astra’s token prices

GPT-6.1 Sol costs significantly less than GPT-6 Astra, but developers should estimate the cost of their specific workload before choosing between them.

Coding Sessions Are Longer and Use More Context. Claude Opus 5.5 Is Built with That in Mind.

This source-based guide explains Coding sessions are longer and use more context. Claude Opus 5.5 is built with that in mind., what the official first-party material supports, and what readers should verify before relying on it.

Introducing Claude Opus 4.7

Anthropic’s latest Opus model is designed for difficult coding tasks, sustained reasoning, higher-resolution vision, stronger output verification, and new cybersecurity safeguards.