AI & LLM Tools
Helpers for working with large language models — count tokens before you send a prompt, or compare embeddings with cosine similarity. All computed locally in your browser.
-
LLM Token Counter
Estimate how many tokens your prompt uses and what an LLM API call will cost. Live character, word and token counts with editable input/output prices — all in your browser, nothing uploaded.
-
Cosine Similarity Calculator
Compute cosine similarity, dot product, Euclidean and Manhattan distance, and the angle between two vectors or embeddings. Paste numbers in any format — runs entirely in your browser.
-
LLM GPU Memory (VRAM) Calculator
Estimate the GPU memory (VRAM) a large language model needs for inference or training, from its parameter count, precision (fp16/int8/int4) and optimizer. Includes an optional KV-cache estimate. Runs in your browser.
-
JSONL Converter & Validator
Convert between JSONL (newline-delimited JSON) and a JSON array, and validate every line with exact line numbers for errors. Great for LLM fine-tuning datasets. Runs entirely in your browser.