sigmoid.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A social space for people researching, working with, or just interested in AI!

Server stats:

591
active users

#prompt

3 posts2 participants0 posts today

fun on an 8 year old Thinkpad (, i7, 32GB mem) with LMStudio and -coder-30B. Impressive 5 to 8 tokens/sec. On my main computer (, 24GB) it's 180 token/sec.

: "Please create a C program that implements a scientific calculator from the command line. Please take special care on the use of paranthesis and correct numbers. Provide the location of errors in the input, when reporting an error."

🧠 Qualche altro test con #Veo 3 "fast" (text-to-video), usando il mio agente #AI per ottimizzare i #prompt
👉 I dettagli: linkedin.com/posts/alessiopoma

___ 
✉️ 𝗦𝗲 𝘃𝘂𝗼𝗶 𝗿𝗶𝗺𝗮𝗻𝗲𝗿𝗲 𝗮𝗴𝗴𝗶𝗼𝗿𝗻𝗮𝘁𝗼/𝗮 𝘀𝘂 𝗾𝘂𝗲𝘀𝘁𝗲 𝘁𝗲𝗺𝗮𝘁𝗶𝗰𝗵𝗲, 𝗶𝘀𝗰𝗿𝗶𝘃𝗶𝘁𝗶 𝗮𝗹𝗹𝗮 𝗺𝗶𝗮 𝗻𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿: bit.ly/newsletter-alessiopomaro 

OpenAI says ChatGPT now handles 2.5B+ prompts daily—that’s over 912B per year 📈🤯
330M of those come from U.S. users 🇺🇸

While still behind Google’s 5T annual searches, ChatGPT’s rapid growth signals a shift in how people seek info 🧠
With an AI browser rumored and Agent tools already live, competition is heating up 🔍

@theverge@mastodon.world
@theverge@flipboard.com

theverge.com/news/710867/opena

Vector illustration of the Open AI logo.
The Verge · OpenAI says ChatGPT users send over 2.5 billion prompts every dayBy Emma Roth

#Prompt engineering for #bibliographic
web-scraping | Scientometrics #openaccess
"how to efficiently use prompts engineering to elaborate a suitable data entry model, able to generate in a single interaction with ChatGPT-4o" from "Bibliographic catalogues"
link.springer.com/article/10.1

SpringerLinkPrompt engineering for bibliographic web-scraping - ScientometricsBibliographic catalogues store millions of data. The use of computer techniques such as web-scraping allows the extraction of data in an efficient and accurate manner. The recent emergence of ChatGPT is facilitating the development of suitable prompts that allow the configuration of scraping to identify and extract information from databases. The aim of this article is to define how to efficiently use prompts engineering to elaborate a suitable data entry model, able to generate in a single interaction with ChatGPT-4o, a fully functional web-scraper, programmed in PHP language, adapted to the case of bibliographic catalogues. As a demonstration example, the bibliographic catalogue of the National Library of Spain with a dataset of thousands of records is used. The findings present an effective model for developing web-scraping programs, assisted with AI and with the minimum possible interaction. The results obtained with the model indicate that the use of prompts with large language models (LLM) can improve the quality of scraping by understanding specific contexts and patterns, adapting to different formats and styles of presentation of bibliographic information.

🧠 #Veo 3, con la nuova funzionalità image-to-video, acquisisce ancora più potenziale. 

✨ Ho generato questi video partendo da immagini generate con #Imagen 4, e pilotando la scena (e i suoni) attraverso dei #prompt testuali. 

💡 L'aderenza alle istruzioni è davvero notevole, come la coerenza con le immagini di partenza. 

___ 

✉️ 𝗦𝗲 𝘃𝘂𝗼𝗶 𝗿𝗶𝗺𝗮𝗻𝗲𝗿𝗲 𝗮𝗴𝗴𝗶𝗼𝗿𝗻𝗮𝘁𝗼/𝗮 𝘀𝘂 𝗾𝘂𝗲𝘀𝘁𝗲 𝘁𝗲𝗺𝗮𝘁𝗶𝗰𝗵𝗲, 𝗶𝘀𝗰𝗿𝗶𝘃𝗶𝘁𝗶 𝗮𝗹𝗹𝗮 𝗺𝗶𝗮 𝗻𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿: bit.ly/newsletter-alessiopomaro 

Im neuen Newsletter #prompt des @iRightsLab dreht sich diesen Monat alles rund ums Thema #KI -Agenten. Was sind sie? Wie funktionieren sie? Wie unterscheiden sie sich von Chatbots? Und wie ist hier die Rechtslage?

Diese Fragen beleuchten wir hier in einem Auszug aus dem Newsletter.
irights.info/artikel/ki-agente

iRights.infoKI-Agenten im FokusKI-Agenten erweitern die Möglichkeiten von KI-Systemen erheblich, doch die rechtlichen Rahmenbedingungen dafür sind noch im Aufbau.

🧠 #Meta e #OpenAI pubblicano nello stesso giorno una guida per la migrazione dei #prompt per i rispettivi modelli. 

👉 I dettagli e la documentazione: linkedin.com/posts/alessiopoma

___ 

✉️ 𝗦𝗲 𝘃𝘂𝗼𝗶 𝗿𝗶𝗺𝗮𝗻𝗲𝗿𝗲 𝗮𝗴𝗴𝗶𝗼𝗿𝗻𝗮𝘁𝗼/𝗮 𝘀𝘂 𝗾𝘂𝗲𝘀𝘁𝗲 𝘁𝗲𝗺𝗮𝘁𝗶𝗰𝗵𝗲, 𝗶𝘀𝗰𝗿𝗶𝘃𝗶𝘁𝗶 𝗮𝗹𝗹𝗮 𝗺𝗶𝗮 𝗻𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿: bit.ly/newsletter-alessiopomar 

prompts

Introducing `prompts`, the latest breakthrough in smoltech AI research. A terminal prompt roulette that will make your panel moderation much more exciting (and cheaper).

git.bleu255.com/prompts/files.

So far `prompts` has been successfully deployed in artistic research panel moderation (true story 🍷). It will be intensively re-trained on a new thematic next week in Brussels to help moderate a digital art panel discussion at @imal for their 25 years anniversary and the launch of the introspective exhibition "I am vertical (but I would rather be horizontal)".

imal.org/en/events/i-am-vertic

doors open at 18:30, starts at 19:00 Tuesday 1st July