After years with Grammarly, I wanted a simpler, cheaper way to improve my writing. So I built Scramble, a Chrome extension that uses an LLM for writing enhancements.
Key features:
- Uses your OpenAI API key (100% local)
- Pre-defined prompts for various improvements
- Highlight text and wait for suggestions
- Currently fixed to GPT-4-turbo
> Key features: - Uses your OpenAI API key (100% local)
Sorry, but we have a fundamental disagreement on terms here. Sending requests to OpenAI is not 100% local.
The OpenAI API is not free or open source. By your definition, if you used the Grammarly API for this extension it would be a 100% local, open source alternative to Grammarly too.
Without marketing speak can I ask why anyone would have a need for a service like grammerly, I always thought it was odd trying to sell a subscription based spell checker (AI is just a REALLY good spell checker).
Non-native speakers find it useful since it doesn't just fix spelling but also fixes correctness, directness, tone and tense. It gives you an indication of how your writing comes across, e.g. friendly, aggressive, assertive, polite.
English can be a very nuanced language - easy to learn, difficult to master. Grammarly helps with that.
I'm a big fan of Grammarly and have been using it, and paying for it, for years.
The advantage is not spell checking. It is grammar and style improvements. It tells you things like "this language is informal", or "this is a better word for that".
The "grammar" part, at least in a professional setting. You might be shocked at how many people will write an email pretty much like they would talk to friends at a club or send a text message (complete with emojis!) or just generally butcher professional correspondence.
It is widely used in countries where the professional language is English, but the native language of the speakers is not.
For example, most Slavic languages don't have the same definite/indefinite article system English does, which means that whilst someone could speak and write excellent English, the correct usage of "a" and "the" is a constant conscious struggle, where having a tool to check and correct your working is really useful. In Greek, word order is not so important. And so on.
Spell check usually just doesn't cut it, and when it does (say, in Word), it usually isn't universally available.
Personally, I have long wanted such a system for German, which I am not native in. Lucky for me DeepL launched a similar product with German support.
A recent example for me was that I was universally using "bekommen" as a literal translation of "receive" in all sentences where I needed that word. Through DeepL I learned that the more appropriate word in a bunch of contexts is "erhalten", which is the sort of thing that I would never have got from a spell check.
I don't think the point here should be the cost, but the fact that you are sending everything you write to OpenAI to train their models on your information. The option of a local model allows you to preserve the privacy of what you write.
I like that.
Assuming for the moment that they aren't saying that with their fingers crossed behind their back, that doesn't change the fact that they store the inputs they receive and swear they'll protect it (Paraphrasing from the Content section of the above link). Even if it's not fed back into the LLM, the fact that they store the inputs anywhere for a period of time is a huge privacy risk -- after all a breach is a matter of "when", not "if".
Makes sense. Strongly hope it won't be a "mac app" but a cross-platform application instead though, nothing worse than having a great mac app that you can't use 50% of the time because your work computer's a mac and your personal computer's a windows machine because you like playing games.
Key features: - Uses your OpenAI API key (100% local) - Pre-defined prompts for various improvements - Highlight text and wait for suggestions - Currently fixed to GPT-4-turbo
Future plans: add LLM provider/model choice, custom prompts, bug fixes, and improve default prompts.
It's probably buggy, but I'll keep improving it. Feedback welcome.
GitHub: https://github.com/zlwaterfield/scramble