“research”

Generates random numbers and random assignment sets for research and classroom work.

An on-device research agent built on LFM2.5 2.6B that runs through WebGPU in the browser.

Chat assistant that pairs the DeepSeek V3 model with live web research.

Shows an AI deep-research transcript with reasoning steps, searches and visited pages.

Pulls related searches recursively to expand a seed keyword into many ideas.

Surfaces People Also Search For keywords ranked by importance.

Shows how identifiable your browser fingerprint is compared with other visitors.

Produces keyword ideas from a seed term for SEO and PPC.

Upscales low-quality images with a FLUX ControlNet model.

Generates custom images from a text prompt combined with one or more reference images.

Shows fresh high-entropy random strings each page load, ready to copy as passwords or keys.

Form that creates a security.txt file telling researchers how to report vulnerabilities.

Suggests keyword ideas with search volume and difficulty for a seed term.

Illustrates how password length and character variety affect brute-force search space.

Click numbered squares in order after they disappear, a working-memory test based on chimp research.

Compiles and runs code in 60-plus languages and keeps a shareable paste of the result.

Searches a dataset of 150M People Also Ask questions for a keyword.

Answers math, science and unit questions typed in plain language, with some features gated behind Pro.

Compares how often search terms are searched over time and by region.

Translates text across 46 languages with a 12B model.

Detects AI-generated text by measuring diversity in how surprising its tokens are.

Ranks a list of keywords by embedding similarity to a primary keyword; OpenAI embeddings are optional.

Generates, edits and answers questions about images and videos with ByteDance's Lance model.