Author here. Jeffy is currently a collection of simple classifiers that can do spam detection, intent routing, topic classification, sentiment analysis and even play Doom
Jeffy works locally on a regular computer, no GPU needed. It comes with 13 already-trained classifiers, and you can also use Jeffy to train your own classifiers
And there’s a lot that can be built on top of this style of architecture
Curious what tasks people find useful, and what architectures you are using for similar tasks
From my perspective. The huge success of LLMs have done two things: 1) shed a lot of light on the inner workings of models and methods, 2) via agents, made it possible for a lot more people to easily build and train their own models
That’s been true for the past 2-3 years, but more recently Typesafe launched Jev with a lot of visibility. This has created quite a bit of hype around the types of tasks that Jev can perform, many of which can be done by a classifier
8 comments
[ 0.22 ms ] story [ 17.8 ms ] threadJeffy works locally on a regular computer, no GPU needed. It comes with 13 already-trained classifiers, and you can also use Jeffy to train your own classifiers
And there’s a lot that can be built on top of this style of architecture
Curious what tasks people find useful, and what architectures you are using for similar tasks
It's a small machine, so it might end up going down. It has some demos of the classifiers within Jeffy, including the Doom player
* Jeffy Playground: https://playground.jeffyclassify.com
* Jeffy home: https://jeffyclassify.com
* GitHub repo: https://github.com/nicobrenner/jeffy/stargazers
That’s been true for the past 2-3 years, but more recently Typesafe launched Jev with a lot of visibility. This has created quite a bit of hype around the types of tasks that Jev can perform, many of which can be done by a classifier