satirr

satirr on Hacker News

Re:Developing Retro-GUI for Win3.11 (Delphi 1) and OS/2 (Visual C++) (... on Sunday November 01, 2026 @05:21PM (#50330535)

What does this mean?

  In the end I’m now able to create applications using the same compiler and code-base for three different architectures (16-bit, 32-bit and 64-bit)

As mentioned earlier, he used the C++ v1.52 (1993) for Windows 3.x. It is a Pure 16-bit compiler. The resulting binary remains NE (16-bit). So, they cant generate a 32 / 64-bit Binary.

Re:HackerTube: YouTube built from videos recommended by HN on Sunday November 01, 2026 @03:36AM (#50323683)

Or also you can use the chrome extension

https://chromewebstore.google.com/detail/dearrow-better-titl...

The creator of this extension is the same as the creator of SponsorBlock. It's more of a crowdsourced.

Re:DualChem Benchmark on Thursday October 29, 2026 @06:43PM (#50297286)

https://fedscoop.com/scale-chemistry-ai-benchmark-caissi-par...

https://satirrdev.github.io/post/?slug=scales-dualchem-bench...

DualChem Benchmark

Posted by satirr on Thursday October 29, 2026 @06:43PM
from the scale.com dept.

https://scale.com/blog/dualchem-evaluating-ai-models-dual-use-chemistry-risks

( Read More... | 1 comment | on HN )

Re:Anthropic opens research preview of Claude Duet, a agent for profes... on Wednesday October 21, 2026 @07:12PM (#50188947)

https://satirrdev.github.io/post/?slug=anthropic-opens-resea...

https://claude.com/resources/articles/claude-duet

Anthropic opens research preview of Claude Duet, a agent for professionals

Posted by satirr on Wednesday October 21, 2026 @07:12PM
from the reuters.com dept.

https://www.reuters.com/technology/anthropic-unveils-claude-duet-ai-agent-app-2026-10-21/

( Read More... | 2 comments | 2 points | on HN )

Anthropic proposes AI safety bill with $1B / 50-death treshold

Posted by satirr on Sunday October 18, 2026 @06:46PM
from the politico.com dept.

https://www.politico.com/news/2026/10/16/anthropic-ai-safety-01124654

( Read More... | 1 comment | 3 points | on HN )

Re:DeepSeek search and fetch – $0.75 per 1000 successful requests on Sunday October 18, 2026 @05:51AM (#50139546)

is the cost of search and scrape provider is impact the quality too? for the search is yes, because there it shows the quality and depth of the search. but for scrape, wht the cons?

Re:Muse Spark Open on Sunday October 18, 2026 @12:13AM (#50137858)

Garmin.img thing is the interesting, but not really on my Experience.

Been running the same stuff on GLM 5.3 Flash, Deepseek v4p, Qwen 3.x model and Mimo v2.x pro model. Mimo faster yeah, but its hallucinate more and losses the thread easier. And when the context window is not limited, when usage of >200k token, is loss context. even he didnt know what im sayin.

Also, sometimes he Had a Loop over and over, for the same tool call (on my case is Grep and Read. >15x looping) until is burning a bunch of context for nothing. GLM and Deepseek doesn't do that (but the deepseek Hallucinations is huge asf too).

Is your Mimo do the same too??

Re:Ask HN: How do you manage skills files? on Tuesday September 08, 2026 @06:45AM (#49608486)

Yes, but isn't that kind of a waste? If there's not enough AI training data or even just the internet, how can we find information just from a command called a skill? It would be nothing?.