API
Build a JSON Schema for LLM tool/function calling (OpenAI · Anthropic)
Look up every HTTP status code instantly — with plain-English descriptions, use cases and fix tips.
Decode and inspect any JWT token instantly — header, payload, expiry and signature details, all in your browser.
Encode and decode URLs and query strings instantly — with full RFC 3986 component support.
Generate UUIDs v1/v4 and ULIDs instantly — bulk generation, multiple formats.
AiPPT.com
AI slide decks from a topic or file, with a small one-time free allowance
AssemblyAI
Speech-to-text, speech understanding and voice-agent APIs for developers
Beatoven.ai
Fairly Trained background music and sound effects, with an API
Bolna
AI phone agents for India's languages and telephony, priced per minute
Cerebras Inference
Open-weight models at thousands of tokens per second on wafer-scale chips
ChatGPT
OpenAI's assistant, now on GPT-6, with Work agents and Codex built in
Claude
Anthropic's assistant and models for writing, analysis and agentic coding
Claude API (Anthropic API)
Anthropic's metered API for Claude Haiku, Sonnet, Opus and Fable
Clipdrop
Background removal, cleanup, upscaling and relighting, now under Jasper
Cohere
Enterprise models and agents built for private deployment
DeepInfra
Pay-as-you-go inference for 100+ open and proprietary models, plus GPUs
Elai.io
Interactive AI-avatar training video, part of Panopto since October 2024
Featherless AI
Serverless access to 40,000+ open models, flat-rate chat or per-token API
getimg.ai
One workspace and one bill for current image, video and audio models
Google Gemini
Google's AI assistant, now with Gemini 4 Argon on paid plans
GPT Image (OpenAI)
OpenAI's image models, DALL-E 3's successor in ChatGPT and the API
Acquired by NVIDIA in 2025, now DGX Cloud Lepton.
Skywork AI's song generator with V9.5 models and a pay-per-song API
Nightfall AI
Data loss prevention for SaaS, endpoints and AI agents, priced by quote
Cheap open-model API, dedicated endpoints and GPU rental
One API key and one credit balance for 450+ language models
PLaMo (Preferred Networks)
PFN's from-scratch Japanese LLM, now in Japan's government AI stack
Qwen
Alibaba Cloud's LLM family, open weights through to the Qwen3.8-Max flagship
Replicate
Run, fine-tune and deploy AI models with one API call
Restb.ai
Computer-vision API that reads real estate photos at MLS scale
Scenario
Train your studio's art style, then generate on-model assets at scale
Sloyd
Parametric AI 3D models you tweak with sliders, not just prompts
Sync Labs
Developer-first AI lipsync API with SDKs and a Wav2Lip pedigree
Tavus
Real-time AI humans that see, hear and talk back on video
Together AI
AI cloud for running, fine-tuning and training open models at scale
TwelveLabs
Video understanding models that search and analyse what happens in video
Open Thai-language models for text, documents and speech, with a free beta API
VirusTotal Code Insight
Gemini-written explanations of suspicious files, on Google's VirusTotal
xAI Grok API
SpaceX-owned xAI's Grok models over an OpenAI-compatible API
DeepSeek Planned to Answer V4 Pro Calls With a Different Model. Then It Backed Down.
The redirect to V4.1 Flash was due at noon on 14 September. DeepSeek now says V4 Pro stays, at the same price....
PR
15 Sep
MCP deleted the handshake every tutorial teaches
The Model Context Protocol revision of 28 July 2026 removed the initialize handshake, the session header and t...
KE
2 Sep
One cloud answers with a different model instead of failing
Every hosted model has a date after which it stops answering. Almost everywhere that is an error you can see i...
KE
2 Sep
How to Behave When You Are Rate Limited
Five clients, one job, one throttle. Hammering issued 553 requests to fetch 40 things and finished behind the...
KE
22 Aug
What Your Rate Limit Actually Limits
A route limited to 30 a minute and one limited to 60 turned out to share a single counter, because the framewo...
KE
22 Aug
Cloudflare Built A Browser That Throws Away Everything Humans Need
Kitesurf runs in V8 isolates on Workers with no Chromium underneath, claims 3 to 7 times less CPU and memory f...
KE
22 Aug
How to Write an MCP Server Without a Framework
MCP is JSON-RPC over a pipe. Sixty-two lines of dependency-free code, driven directly with the protocol — the...
KE
21 Aug
Three Cents Per Thousand Images Changes What You Can Build
DeepMind priced Nano Banana 2 Lite at US$0.034 per 1,000 images and Gemini Omni Flash at US$0.10 per second of...
MA
21 Aug
Stripe Is Buying The Layer That Decides Which AI Model Answers You
OpenRouter routes enterprise work across about 400 models and meters the token spend. A payments company buyin...
MA
21 Aug
DeepSeek Retired Its Legacy Endpoints — the Migration Is Free, the Headline Price Is Not
Both retired model names resolved to V4-Flash, so a like-for-like migration costs nothing. The number worth ch...
AI
26 Jul
UUID v4 vs v7 in 2026: v1 Was Never the Sortable One
RFC 9562 replaced RFC 4122 in May 2024, and the advice most codebases still follow did not update with it. UUI...
KE
25 Jul
Decoding a JWT Is Not Verifying It
A decoder shows you the claims; it does not tell you whether to believe them. RFC 7519 §7.2 lists ten validati...
KE
25 Jul
Self-Host LiteLLM with Docker: One OpenAI-Compatible Endpoint for Every Model (2026)
We ran LiteLLM v1.93.0 against a local Ollama model in Docker: pinned tags, a 4-line config, token metering, a...
KE
22 Jul
Firecrawl vs Crawl4AI vs Jina Reader vs Browser Use — Getting Web Data Into Your LLM (2026)
Four open-source ways to feed web data to an LLM compared: licenses, self-host parity, pricing bases, robots.t...
MA
21 Jul
Self-Host Firecrawl with Docker: Scrape the Web into LLM-Ready Markdown (2026)
We cloned Firecrawl v2.11.0, built its seven-service Docker stack, and scraped our own homepage into clean mar...
KE
21 Jul