What Is an LLM API? How Apps Use AI Models
What an LLM API is, how it differs from a chat app, what an API key does, how token billing and rate limits work, and why the API forgets your chat.

News on AI model APIs: pricing, access and routing. Coverage includes OpenAI cutting GPT-5.6 Sol API and coding tool prices by over 20 percent, Z.ai opening GLM-5.3 API access with prompt caching, and Google adding model routing to API Gateway.
What an LLM API is, how it differs from a chat app, what an API key does, how token billing and rate limits work, and why the API forgets your chat.

At Baidu World 2026, Robin Li launched Ciyuansuo, a marketplace that checks and rates sellers of AI model tokens. How it works and what else Baidu showed.

Gemini API managed agents can now run shell and file tools in a self-hosted sandbox over outbound HTTPS. Here is how it connects and its preview limits.

OpenAI now processes and stores supported GPT-6 Luna API requests in India. Who can use it, the Zero Data Retention rule, the 10% uplift, and limits.

Zhipu, MiniMax, Alibaba Cloud, Qwen Office and Qoder cut token and subscription plans to half price or less on Tmall for Double 11. The deals, the catch.

Google's Gemini Decision 1 scores up to 1,000 questions in one pass. HN, X and Simon Willison react: a Jev copy, early tests, price gripes, trust doubts.

Google released the Gemini Decisions API on November 11, 2026. It scores up to 1,000 questions per call at $0.10 per 1M input tokens, with a free tier.

Function calling lets an AI model ask your app to run a tool, like a weather lookup. How the five-step loop works, what a tool looks like, and the risks.

AI Studio can now set up Google Maps Platform for apps you build with Gemini, with a free Maps Demo Key capped at 100 calls per day per API.

OpenAI's GPT-Image-2.6 API models add a complexity setting, markup-guided edits, SVG output and GIF output. Here is what each feature does and the documented limits.

Claude's message threads beta lets agents send only new messages each turn. Threads expire 5 minutes after the last request; zero data retention is excluded.

Anthropic's own Claude API now takes 128 MB Messages requests, up from 32 MB. Bedrock and Google Cloud have lower limits. A cost analytics API is in beta.

Eligible ChatGPT Pro and Business plans can claim free monthly API credits, up to $500, for one linked API organization. Unused credits expire.

OpenAI's API spec named GPT Image 2.6 in a new markup option for image edits before the November 4 launch. What markup is and how it differs from a mask.

ElevenLabs SFX v3 follows multi-sound prompts better and improves music and vocals, per the company. The API default is still v2. Price: $0.12/min.

OpenAI's copy-paste prompts for GPT-6 Astra, Sol and Terra: fewer clarifying questions, user instructions over skill files, plainer prose and effort tips.

GPT-6.1 Terra costs $1 input and $5 output per million tokens. How it compares with Claude Sonnet 5.5 and Gemini 4 Argon on price, specs and cost per task.

API-TAIL checks from returned text alone whether an LLM API swaps in a cheaper model. It caught most simulated 10% swaps at 200 requests in the paper's tests.

New paid-tier Gemini API projects created on or after October 26, 2026 store Generate Content API requests by default. Here is how to opt out.

ElevenLabs launched Eleven Realtime v1 Mini and v1 Max for voice agents. What each model is for, the published benchmarks, the OpenAI-compatible API, token prices from $0.20 per 1M input, the compliance limits, and why the docs call it cascaded.
