Examples
Each page below collects requests of one kind. Every request is runnable: change the fields, press Run, and the response appears under it. Until you run one, each block shows a real response recorded from a vid2context server, and the same request as curl, JavaScript, Python and, where they exist, the CLI command and the MCP tool call.
Runnable requests
- Transcripts: Runnable transcript requests for the vid2context HTTP API: plain text, timestamps, SRT, WebVTT, Markdown and JSON, languages, translation, sponsor removal and captions-only mode.
- Video structure: Runnable requests for a video's structure: metadata, outlines with token counts, single chapters and time ranges, comments, caption languages, SponsorBlock segments, generated chapters and key moments.
- Many videos: Runnable requests for many videos at once: expand playlists, build one context pack for several videos, download every format as a zip, research a topic and search the library.
- Collections and library: Runnable requests for vid2context collections and the saved-video library: create a collection, add videos, read it, pack it, export Obsidian notes and the definition, list and read saved videos.
- Answers, briefs and extraction: Runnable LLM requests over video transcripts: cited answers, streamed answers, cited briefs as background jobs, and structured extraction with templates or your own JSON Schema.
- Your own files: Runnable requests that upload an audio or video file to the vid2context HTTP API and transcribe it, or import a transcript you already have, with a public-domain Apollo 11 clip you can send from the page.
- Library playground: Run the vid2context core library in your browser on a real VideoContext: render every output format with its options, chunk, count tokens, clean the transcript and write your own code.
- Server and models: Runnable requests that describe a vid2context server: its health and version, a deep readiness check, the output formats, every model rung by tier, and the OpenAPI document.
The Server menu on any block picks where requests go: the API on this site, a server on your machine (http://127.0.0.1:8787), or any other URL. The Tutorials overview explains how to start a local server that this page may call.
For step-by-step lessons that use these requests in order, start with Your first transcript.