tynmann
Re:Animated a page with an old 90s look How can I make it more interes...
on Sunday December 10, 2023 @09:43PM (#38597098)
Thank you, this is very good advice.
Re:Animated a page with an old 90s look How can I make it more interes...
on Sunday December 10, 2023 @09:51AM (#38591860)
I wrote a small amount of JS and CSS to give this page: https://thelibrarian.cc an old UI look and animate it in the same style as typical 90s UIs I have seen during my childhood in La Femme Nikita, Unreal Tournament, Starcraft and The Matrix.
It's a landing page for a book I wrote.
What can I do to make it more interesting?
An option I am considering is allowing user interaction after the main animation finishes. So one can explore and browse the files of other subjects and learn more about the corporation. Would you like something like this?
Re:Show HN: Chat with Zephyr 7B over voice message using whisper.cpp a...
on Monday October 16, 2023 @04:12AM (#37896998)
However if one wants to run something like this in production, it would make more sense to use a server with at least an A10, which has better performance per power consumption.
Re:Show HN: Chat with Zephyr 7B over voice message using whisper.cpp a...
on Monday October 16, 2023 @04:09AM (#37896979)
I'm running everything on my own PC. With a 3060 12 GB at under $300, it's very cheap to build a system capable of running inference with smaller models.
But even bigger models (13B or 30B) can be handled with a dual GPU setup (24 GB total memory at < $600), with quantization.
The Discord bot works as a client (outgoing connection to the Discord public server), so it can run from behind NAT.
Show HN: Chat with Zephyr 7B over voice message using whisper.cpp and llama.cpp
https://bastianai.com/blog/2023/10/chat-with-self-hosted-llm-via-discord-voice-message-transcription
Just my small weekend project, experimenting with LLMs. I made a Discord bot that uses the Zephyr 7B model to reply, and whisper.cpp to transcribe voice messages into text. So you can send it voice messages and it replies.
Re:Ask HN: Show me your half baked project
on Thursday October 12, 2023 @02:57PM (#37861475)
Thank you! I pushed a fix for the Firefox issue.
And here is the link to the demo for convenience:
Re:Ask HN: Show me your half baked project
on Thursday October 12, 2023 @12:34PM (#37859475)
I'm working on an LLM-based tool to assist with fiction writing. The goal is to help writers get unstuck in case of writer's block, not to do autonomous writing.
One has to enter some abstract of what they want to write, and some text they have so far, and the tool generates possible continuations.
It requires quite a bit of editing, maybe 3-4 attempts for each completion if one wants to generate decent text. Feedback was mostly positive from a dozen or so professional writers who tried it out for a day.
I feel that it generates more naturally looking text than ChatGPT with the typical prompts.
See my profile if you want to try it out. You can use the demo without registration.
Re:Ask HN: Who wants to be hired? (September 2023)
on Saturday September 02, 2023 @05:55PM (#37365872)
Location: Europe
Remote: yes
Timezone: late hours EU, morning US
Tech: Python, Go, C, Linux. Versatile, can do almost everything except frontend/mobile app development
Email: [email protected]
Software engineer with 6 years of experience, recently backend and before, networks and embedded system programming. I'm very good at what I do and have a couple of top companies on my CV.
Looking for a contract or part time job for a few months to complement my income.