API Updates

API changes, new endpoints, and developer tools.

Last Week

OpenAIapi

Link·Aug 21, 2026

Usage and Spend Tracking by API Key

OpenAI's usage dashboards now break spending down by individual API key, and you can set monthly limits that cut off traffic when reached.

Best for: finding which app or workload is driving an API bill
When to use it: several projects share one OpenAI account and the invoice does not tell you who spent what
Who benefits: whoever owns the AI budget on a team running more than one thing on OpenAI

Perplexityapi

Link·Aug 20, 2026

41 Models in the Perplexity Agent API

Perplexity's Agent API now reaches 41 frontier models from nine providers through a single endpoint, with search and code execution built in.

Best for: building an agent that routes different jobs to different models without wiring up each provider
When to use it: you want model choice but not nine sets of credentials and nine SDKs
Who benefits: developers building multi-model agent workflows

OpenAIapi

Link·Aug 20, 2026

Prompt Caching Dashboard

A dashboard showing how much of your API traffic hits the prompt cache, for developers on the OpenAI API.

Best for: Finding out whether the caching you set up is saving money.
When to use it: API spend is climbing and you suspect the cache is missing more than it hits.
Who benefits: Developers and whoever owns the API bill.

OpenAIapi

Link·Aug 20, 2026

Transparent Backgrounds for GPT-Image-2 (Preview)

Image generation can return a transparent background instead of a filled one, for developers using the API.

Best for: Generating a logo, product cutout, or icon you can drop onto any background without cutting it out by hand.
When to use it: The same generated asset has to sit on a light page, a dark page, and a social card.
Who benefits: Marketing and design teams generating reusable assets through the API rather than one at a time.

Anthropicapi

Link·Aug 20, 2026

Computer Use, Browser Tool, Skills API, and Files API (Generally Available)

Four building blocks for agents on the Claude Platform move out of beta, for developers building on the API.

Best for: Building an agent that has to operate software with no API behind it, like an internal tool or a vendor portal.
When to use it: A workflow only exists as clicks in someone's browser and you want Claude to run it end to end.
Who benefits: Developers and technical teams automating work that lives in applications nobody can integrate with.

OpenAIapi

Link·Aug 19, 2026

Private Safety Processing (Preview)

A system that lets OpenAI spot misuse across related conversations without staff seeing the content, for enterprise and API customers on zero data retention.

Best for: Keeping a zero-retention agreement in place on frontier models instead of trading it away for safety monitoring.
When to use it: Legal or security blocked a frontier model because the safety tooling required keeping data.
Who benefits: Enterprise and API buyers in regulated industries who negotiated zero data retention.

Anthropicapi

Link·Aug 19, 2026

Claude Managed Agents Update (Sandbox Memory and Domain Controls)

Two changes to Claude Managed Agents: memory in self-hosted sandboxes, and limits on what an agent is allowed to search and read.

Best for: Running a managed agent on your own infrastructure without losing what it learned between runs.
When to use it: An agent needs to carry context forward, or it should only reach sources you approve.
Who benefits: Teams running Claude Managed Agents under compliance or data-boundary requirements.

Perplexityapi

Link·Aug 18, 2026

Perplexity API MCP Server as an Agent Plugin

Perplexity's API connector now installs as an Agent Plugin in coding tools like Codex, Cursor, and VS Code.

Best for: Giving a coding assistant live web search without wiring the connection yourself
When to use it: Your assistant keeps answering from stale training data and needs current sources
Who benefits: Developers running agents inside Codex, Cursor, or VS Code

August 2026

Perplexityapi

Link·Aug 14, 2026

Perplexity Search SDK

A Python toolkit that lets an AI agent run and refine its own web searches in code instead of making one search call and taking what comes back.

Best for: building an agent that researches a topic across many sources before it answers
When to use it: a single search call is too blunt and you want the agent to run several searches, then sort and trim the results itself
Who benefits: developers building research or monitoring agents on Perplexity

Perplexityapi

Link·Aug 13, 2026

Sonar Moving to the Agent API

Perplexity is folding its Sonar search API into the Agent API, which keeps grounded web search and adds tools and model choice.

Best for: teams already calling Sonar for grounded web search inside their own products
When to use it: you are starting a new Perplexity API build, or your Sonar calls need more than one search-and-answer step
Who benefits: developers running Perplexity search in production