News for hackers. Stuff that matters.

vincelt

vincelt on Hacker News

Re:Google Antigravity on Tuesday November 18, 2025 @05:10PM (#45972922)
It's 2025 and a code editor is now 600MB.
Re:I'm Making a Beautiful, Aesthetic and Open-Source Platform for Lear... on Monday September 08, 2025 @09:23AM (#45167941)
Agreed. For immersing at home, reading mangas/webtoons with an OCR translated layer and watching Japanese vlogs with dual subtitles has been effective for me.
Re:Agent Client Protocol on Tuesday September 02, 2025 @07:04AM (#45101406)
Found this repo the other day in case that's useful: https://github.com/Xuanwo/acp-claude-code
Re:Gemini CLI on Thursday June 26, 2025 @02:43AM (#44384792)
Agreed. If there are provider-specific differences then we should be able to add CLAUDE/GEMINI.md to take precedence, but most things in there are basic or general information and should apply to all providers (same for .cursor/, .windsurfrules/, ..)

Show HN: Language Learning with Comics

Posted by vincelt on Thursday April 03, 2025 @12:59PM
from the show-hn dept.

https://lexirise.app

Hey friends,I'm currently learning Mandarin and was looking for a way to practice my reading that I'd be looking forward to do every day. Graded readers are quickly boring, and normal books are way too far up for my level.Comics and webtoons are the perfect format: fun, conversational style, contains slang. But unfortunately are not compatible with traditional language learning apps (linq,...

( Read More... | no comments | on HN )

Show HN: Abso – TypeScript LLM Client

Posted by vincelt on Wednesday February 12, 2025 @09:40AM
from the show-hn dept.

https://github.com/lunary-ai/abso

Hi HN,I made a TypeScript library that unifies queries with different LLM providers like Anthropic and Gemini into the OpenAI format.When playing around with new models like the new Gemini 2.0, it's frustrating to adapt to the differences in things like streaming, tool calling, json formatting, output formats, etc.This library aims to make it easy to switch between providers, while keeping the...

( Read More... | no comments | on HN )
Re:GPT-4.5 or GPT-5 being tested on LMSYS? on Monday April 29, 2024 @01:55PM (#40201686)
Still fails to solve this one:

Prompt:

> Complete the following Python program:

```

len, print = print, len

def print_len(x):

     "Print the length of x"
```

gpt2-chatbot:

```

len, print = print, len # Swap the functionalities of print and len

def print_len(x): "Print the length of x"

    len(x)  # This 'len' is actually 'print' because of the swap above
print_len("Hello, world!") # This will output: 13

```

Show HN: Open-source observability for LLM apps

Posted by vincelt on Saturday October 28, 2023 @08:58AM
from the show-hn dept.

https://github.com/llmonitor/llmonitor

Hi HN,Hugh and Vince here.LLMonitor helps you record, trace & search your LLM queries and chatbot conversations.You can also capture user feedback on your frontend and correlate it with backend LLM queries then use that to fine-tune your own models.The project started has an internal tool in our previous (failed) AI startup.We’re aware the LLM observability space is very crowded. Apart from...

( Read More... | 1 comment | 4 points | on HN )
Re:Comparing 60 LLMs with a set of 20 prompts on Saturday September 09, 2023 @01:10PM (#37447630)
Have you tried replicating via the API with a temp of 0?
Re:Comparing 60 LLMs with a set of 20 prompts on Saturday September 09, 2023 @01:05PM (#37447576)
Oh neat, thanks for sharing, wanted to add an interpreter to that test
Re:Comparing 60 LLMs with a set of 20 prompts on Saturday September 09, 2023 @12:57PM (#37447457)
I used the API from Together[0].

Thanks for sharing your results, they're indeed pretty different. I looked at the source again and did append a "# " before every prompt made by those 10 `code` models (during testing thought that formatting it as a Python comment might help them).

Will re-run the script without that to see if it matches your results.

[0] https://docs.together.ai/docs/models-inference#code-models

Re:Comparing 60 LLMs with a set of 20 prompts on Saturday September 09, 2023 @10:25AM (#37445811)
Thanks. I haven’t put it online yet, but will try to clean it (removing API keys & all) tonight/tomorrow and publish it
Re:Things I tried to improve sleep on Friday August 25, 2023 @03:57PM (#37266318)
Interesting, what about it makes you sleep better?
Re:Things I tried to improve sleep on Friday August 25, 2023 @03:53PM (#37266274)
Indeed I gave up almost completely alcohol. Will enjoy a few beers at gatherings every 2 months or so but in full consciousness that I’m sacrificing the following night.
Re:Things I tried to improve sleep on Friday August 25, 2023 @07:45AM (#37260609)
Corrected the post and bad math sorry! 1. ~10h is ~8.5h indeed + time to fall asleep (not counted in the % awake on Apple Watch) 2. Curious as well. 3. For me sleeping less than 7h30, I'll definitely feel it the next days. But I'm early 20s and I guess it varies a lot. Good question about recommendations & efficiency,I always wondered about that.
Re:Things I tried to improve sleep on Friday August 25, 2023 @07:41AM (#37260580)
Thanks, never thought about that, I'll sometimes wake up in the middle of the night with a sort of claustrophobic feeling and need to take a deep breath when there is no ventilation or windows open. Girlfriend always complain that I must leave everything open because of that.

Do you use a Co2 meter to monitor?

Older »