News : ... test with the Tascade agent using custom data.

Today I tried to build an agent in taskade using my data.

I wanted to see if I could create SVG files using the data I entered.

I used a link and this PDF file as agent training data.

After training, I asked them to create an SVG file for me.

Here is the result:

FASM : simple bootloader source code with menu .

This is the source code with a boot menu and can be tested with qemu emulator.

I need to use the Sonet 3.5 artificial inteligence to search infos and make this source code better, but I learn assembly in the past.

The compile commands of the source code:

mythcat@localhost:~/fasm$ ./fasm.x64
flat assembler  version 1.73.32
...
mythcat@localhost:~/fasm$ ./fasm.x64 bootloader.asm bootloader.bin
flat assembler  version 1.73.32  (16384 kilobytes memory, x64)
2 passes, 512 bytes.

The test with qemu tool:

mythcat@localhost:~/fasm$ qemu-system-x86_64 -drive format=raw,file=bootloader.bin

Let's see the source code:

; Simple Bootloader with Menu for QEMU
; Written in FASM

org 0x7C00

format binary

start:
    ; Clear the screen
    xor ax, ax
    mov ds, ax
    mov es, ax
    mov ah, 0x0
    int 0x10

    ; Display the menu
    call display_menu

main_loop:
    ; Wait for key press
    call wait_for_input

    ; Check key pressed
    cmp al, '1'
    je handle_option1
    cmp al, '2'
    je handle_option2

    ; Invalid option, restart
    jmp main_loop

handle_option1:
    ; Option 1 selected
    mov si, option1_msg
    call display_message
    call wait_for_input  ; Wait for any key
    jmp start           ; Return to menu

handle_option2:
    ; Option 2 selected
    mov si, option2_msg
    call display_message
    call wait_for_input  ; Wait for any key
    jmp start           ; Return to menu

display_menu:
    ; Display the menu
    mov si, menu
    call print_string
    ret

wait_for_input:
    ; Wait for a key press
    xor ax, ax
    int 0x16
    ret

print_string:
    ; Print string at DS:SI
    mov ah, 0x0E
@@:
    lodsb
    test al, al
    jz @f
    int 0x10
    jmp @b
@@:
    ret

display_message:
    ; Print message at DS:SI
    call print_string
    ret

menu db 'Simple Bootloader Menu', 13, 10
     db '1. Option 1', 13, 10
     db '2. Option 2', 13, 10
     db 'Select an option: ', 0

option1_msg db 13, 10, 'You selected Option 1!', 13, 10
           db 'Press any key to return to menu...', 13, 10, 0

option2_msg db 13, 10, 'You selected Option 2!', 13, 10
           db 'Press any key to return to menu...', 13, 10, 0

rb 510-($-$$)
db 0x55, 0xAA

This is the result:

News : stablecog online tool generate artificial photos.

This news this online tool is old ...

My results with the free version were not very promising. Obviously, the basic test for those who know the theory and even for those who don't is to make substantial changes while keeping the information correctly framed. Obviously it is possible, but I suspect that it is a deepfake protection. More simply, I tried to modify a photo with a person ... if I modify the settings to keep realism, then it does not make any modifications.

I like that it contains many generators with many models of artificial intelligence and can be tested and the free version has a significant number of attempts.

Black Forest Labs, a company with a team that contributed to the creation of the original Stable Diffusion have released their new model called FLUX.1. It is a 12 billion parameter rectified flow transformer. Based on discussions on Reddit and HuggingFace, it seems like the community is very excited about this new model. We are as well, in fact, we think it is the best open-source AI image model currently available. Let’s take a look at what it offers.

News : the Google Research with AI Colab and more ...

I saw few days ago the new AI implementation from Google Colab and I found the Google Research ...

This AI seams to work with development task, not all AI for development works well ...

See more about this on the Google Research official blog ... and the youtube channel - GoogleResearch/videos.

Online tool : circuito.io with artificial intelligence.

... circuito.io is an online platform, generating wiring and code for Arduino projects.

Run Large Language Models Locally

AI should be open source, transparent, and available to everyone. GPT4All enables anyone to run open source AI on any machine.

You can find it on this website.

News : Introducing Writer for Slack

Online tool to animate drawings.

Upload a drawing of one character and transform it into a fun animation using AI.

You can find this online tool on this webpage.

LivePortrait : Add mimics and lip sync to your static portrait driven by a video ...

You can tst online on the huggingface.co - KwaiVGI - LivePortrait

The source code can be found on this GitHub repo.

A demo real video from theAIsearch:

GitHub : Single Images with Multi-View Pose Calibration ...

On this GitHun repo you can find the CharacterGen: Efficient 3D Character Generation from Single Images with Multi-View Pose Calibration

Testing 2D texture with runwayml artificial intelligence.

... as expected, because the artificial intelligence with a dataset to a model specially trained on 2D textures, the result after the request is one according to the dataset.

I have not found a dataset available online for generating 2D textures in the graphics domain for texturing 3D objects.

NOTE : If you want to generate a 2D texture for an airplane to be integrated into a 3D model, then you need a specific dataset.

This result is tested with runwayml by asking:

generate a 2D texture for a plane to use in an ice wall 3d object

... and result is:

News : The Greatest AI Video EVER ... from Matt Wolfe youtube channel.

... a video from Matt Wolfe about the new Gen-3 Alpha on Runway ...

Online claude artificial intelligence

You can test on the official webpage and comes with free and paid version and no A.P.I.

See the answer of this question:

What are your ability ?

As an AI language model, I have a wide range of capabilities. Here's an overview of some key abilities:

1. Information and knowledge:

• I can provide information on a vast array of topics, drawing from my training data up to April 2024.
• I can explain complex concepts in simpler terms.
• I can answer questions on history, science, culture, current events, and many other subjects.

1. Language processing:

• I can communicate in multiple languages.
• I can assist with translations, grammar, and language learning.

1. Analysis and problem-solving:

• I can analyze text, data, or situations provided to me.
• I can help break down complex problems into steps.
• I can assist with logical reasoning and critical thinking tasks.

1. Writing and creativity:

• I can help with various forms of writing, from essays to creative fiction.
• I can brainstorm ideas and assist in creative processes.

1. Mathematics and computation:

• I can perform calculations and help with mathematical problems.
• I can explain mathematical concepts and provide step-by-step solutions.

1. Programming and technical tasks:

• I can write code in various programming languages.
• I can explain programming concepts and help debug code.
• I can assist with technical documentation and explanations.

1. Task planning and organization:

• I can help break down projects into manageable steps.
• I can assist with creating schedules or to-do lists.

1. Data interpretation:

• I can help analyze and interpret data sets or statistics you provide.

1. General assistance:

• I can help with proofreading and editing.
• I can provide summaries of long texts.
• I can engage in discussions on various topics.

Remember, while I have broad capabilities, I also have limitations. I can't access the internet, I don't have real-time information beyond my training data, I can't learn or remember information from our conversation, and I can't perform actions in the physical world.

Is there a specific area you'd like to explore further or a particular task you need assistance with?

Puter online tool with computer operating system.

Puter is a privacy-first personal cloud to keep all your files, apps, and games in one secure place, accessible from anywhere at any time.

You can find it on this website.

In the terminal use this command :$ help

You can have an artificial inteligence:

$ ai "tell me a poetry"
In the world of bits and bytes,  
I dwell within the screen's light.  
A Puter AI, full of might,  
Ready to assist and guide your way,  
With commands you enter each day.

You can run this in Minecraft game ...

News : rendernet.ai

Another online tool with AI, see more on this webpage.

The account is @walkingcat_9015, how funny !

News : Speaking online tool !

Here is an online tool that allows you to complete a speech in both audio-to-text, predefined or custom voices ...

See more on the elevenlabs.io official website.

SayMotion use artificial intelligence to create animations.

SayMotion™ by DeepMotion brings this vision to life, going beyond other generative 2D AI tools. With a simple text prompt (video prompt coming soon!), intricate 3D motions are generated effortlessly in minutes, allowing creators to take the director's seat and bring digital actors to life with their words. Sign up for the SayMotion Open Beta: https://www.saymotion.ai

Craft & Train AI Agents with Advanced Knowledge — Build and Activate Custom Commands in Taskade AI

Suno AI - Artificial intelligence tool for creating songs.

You can find on this website.

Online walkthrough of the GPT large language models.

A visualization and walkthrough of the LLM algorithm that backs OpenAI's ChatGPT. Explore the algorithm down to every add & multiply, seeing the whole process in action.

The visualization works like an online presentation and can be found on this webpage.