What it does
A word-level language model that drafts with n-grams and asks Jev to check chunks of the result.
Benchmarks & research
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
A word-level language model that drafts with n-grams and asks Jev to check chunks of the result.
You can use this experiment to explore how software chooses words to complete sentences. Start by writing down a few short opening lines to test. A developer can connect the tool with an access key from TypeSafe to show how it drafts and verifies words.
Treat this as a learning project rather than a working writing tool. It recognizes only 229 words, costs thousands of service credits per word, and writes worse than simple word counters. Because of these harsh limits, do not use it in real products.