APIs and SDKs provide the programming interfaces and software kits developers use to integrate artificial intelligence models into applications. Truncated covers new interface releases, protocol updates, developer tools and integration standards.
AI-generated summary OpenAI hosted a workshop titled New Ways of Working with Codex at the AI Engineer Code Summit. OpenAI Developer Experience engineers Dominik Kundel and Charlie Guo demonstrated new Codex features and voice mode capabilities.
AI-generated summary OpenRouter explains how API rate limits and 429 errors occur when using OpenAI, Anthropic, and OpenRouter. OpenRouter details fixes including waiting for reset times, using fallback models, and routing requests through different providers.
AI-generated summary Google released Gemini 4 Carbon to everyone after its initial announcement for cyber defenders two weeks ago. The model costs two dollars per million input and ten dollars per million output at launch before a price increase on December 1st.
AI-generated summary Latent Space discussed agent experience with Resend co-founder and CEO Zeno Rocha on its latest podcast episode. Resend says it had 4.3 million users and 120,000 paying customers at the time of the recording.
AI-generated summary Google details on-device speech recognition and text-to-speech features available in the Google AI Edge Gallery for developers. The open-source code on GitHub provides optimized solutions using ML Kit, MediaPipe Tasks, and LiteRT for Android and cross-platform applications.
AI-generated summary OpenRouter explains how to obtain reliable structured output from any large language model using JSON schemas, tool calls, or decision models. OpenRouter says users can configure strict schemas, use response healing, and validate every reply to prevent parsing failures and code breaks.
AI-generated summary Unite.AI published an editorial evaluation examining ten artificial intelligence video generation application programming interfaces for developer workflows in November 2026. The guide highlights platforms including Runway, Google, BytePlus Seedance, and Luma for their capabilities in generation, editing, and reference handling.
AI-generated summary Simon Willison has added support for the newly released Anthropic Fable 5.2 and Mythos 5.2 models to llm-anthropic. The author noted that the max reasoning level failed to return an image, running out of tokens during testing.
AI-generated summary Amazon Web Services announced that Amazon Bedrock Managed Knowledge Base now supports single-shot retrieval-augmented generation and passage-only requests through AgenticRetrieveStream. AWS says single-shot operation can reduce latency and model usage relative to agentic retrieval, though the announcement supplies no measured latency or cost comparison.
AI-generated summary The support notice explains that the Messages API handles thinking blocks differently to protect against distillation. The change affects models including Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 5.5, Claude Haiku 5.5, Claude Fable 5.2, and Claude Mythos 5.2.
AI-generated summary Paid Claude plans provide a standard context window size of 200K tokens for chatting with Claude or using Claude Code and Claude Cowork. A larger 1M context window is available by selecting specific models through the command line when usage credits are enabled.
AI-generated summary LangChain has released Promptim, an experimental open-source library designed to help developers systematically improve AI systems through automated prompt optimization. The library integrates with LangSmith to support dataset management, tracking results, and incorporating human feedback during the optimization loop.
AI-generated summary Google announces that developers can now use Google Maps Platform directly within Google AI Studio. The integration allows developers to connect real-world location data to Gemini and load a live interactive preview.
AI-generated summary Octomind examined Claude API pricing across models including Claude Haiku 5.5, Sonnet 5.5, Opus 5.5, and Fable 5.1 based on Anthropic documentation checked in November 2026. Anthropic reports rates per million tokens and outlines prompt caching, batch processing discounts, and model lifecycles for its Claude API offerings.
AI-generated summary Octomind version 0.70.1 introduces a single setup screen that prompts users to select an LLM provider upon initial startup. The update also includes idle context compaction designed to shorten conversations when provider caches expire.
AI-generated summary LG AI Research announced free API access to K-EXAONE 2.5 on November 9, 2026, offering hosted use of the 750-billion-parameter model through the EXAONE API Platform. LG says developers, startups, companies and research institutions can obtain an API key to use the text-only model supporting Korean and English for service development and research.
AI-generated summary Forkast reports that the OpenAI Agents SDK alpha update changes environment execution defaults, Model Context Protocol approval configuration, and bundled reasoning defaults. The pre-release notes specify new write policy requirements, renamed approval strings, and adjusted reasoning effort levels for three named models.
AI-generated summary Google ADK v2.14.0 updates RunSkillScriptTool in environment mode to replace free-form shell commands with structured arguments and path validation. Google states this change closes a hole that allowed a prompt-injected model to run arbitrary shell commands.
AI-generated summary The user shares a personal use case for Teammate involving Perplexity people-search-sdk to identify candidates from headhunter descriptions. The workflow filters profiles for internal needs and drafts outbound messages to remove the middleman.
AI-generated summary Fireworks AI VP Rob Ferguson explains that using a portfolio of models and automated routing through Nexus and its API helps reduce costs and improve utilization. Ferguson also notes that caching rates and data retention policies impact overall expenses while platforms like Microsoft help handle sensitive workloads.
The top AI news in your email delivered each morning. Choose categories