Monday, November 4, 2024
Show HN: Krixik – Easily sequence small/specialized AI models (pip-installable) https://ift.tt/7YIkjGy
Show HN: Krixik – Easily sequence small/specialized AI models (pip-installable) Our initial intention was to experiment with a smartbook idea, but we couldn't find a good prototyping tool for small/specialized AI models. That rabbit hole took us through vector databases, through infra for small AI, and finally here. We’re particularly proud of the accessibility/simplicity of our code syntax. Krixik’s model library is limited to eleven model types (modules) and a few dozen models, but we will significantly expand it. Other enhancements, like a local client and several 3rd-party integrations, are also planned. You require API credentials to try Krixik. You can quickly get them through this form: https://ift.tt/f0Khbn3 And here’s a demo video (our YouTube channel also has a couple dozen example videos in it): https://youtu.be/WpSSYLfvfdM https://ift.tt/gcrE5tP November 4, 2024 at 07:03PM
Sunday, November 3, 2024
Show HN: I wrote a techno-thriller on AI and need feedback (no signup) https://ift.tt/7lBw5xe
Show HN: I wrote a techno-thriller on AI and need feedback (no signup) https://ift.tt/EvdcQi4 November 3, 2024 at 08:44PM
Saturday, November 2, 2024
Show HN: GraphQL Zeus 7 – type-safe GraphQL on front end for newbies https://ift.tt/rf8HIph
Show HN: GraphQL Zeus 7 – type-safe GraphQL on front end for newbies Ok, so I've heard the voice of community and added null support and dropped const enums. Also you can fetch all scalar fields with "fields" selector. https://ift.tt/MiE25sl November 2, 2024 at 05:45PM
Friday, November 1, 2024
Show HN: I made an interactive sentiment model comparison site https://ift.tt/9gcrL7j
Show HN: I made an interactive sentiment model comparison site Hey HN, I needed to assess the state of sentiment models and couldn’t find a good way to compare them. I built this interactive site that lets you compare 12 models side by side, from Python libraries like NLTK Vader, to top performing models on HuggingFace, to commercial sentiment APIs and GPT4o. This is a research project, there is no paywall - you can enter your own text (1) and get the results back immediately. The results are fascinating and we made it easy to explore not just the leaderboard, but where models get it wrong. For example, most models (including AWS Comprehend) can’t get this positive sentiment: "Food doesn’t get better than this. I was sad when I finished, actually sad. To die for." (2) And yes, GPT4o is currently the best performing. It's crazy how many laboriously researched models are superseded by general purpose foundation models. Let me know what you think? 1: https://ift.tt/jVE76eM 2. https://ift.tt/1Cl7ReG... https://ift.tt/duJfKWQ November 1, 2024 at 11:49PM
Subscribe to:
Posts (Atom)