Thanks, though I will not install it. I like llama.cpp as it is straghtforward, npm packaging is hell.
It is proprietary
Meta’s LLaMa 2 license is not Open Source – Open Source Initiative:
https://opensource.org/blog/metas-llama-2-license-is-not-open-source
I don't care for META, I am talking to you, when you release proprietary LLM...
There are consequences which you did not think of.
What if you really make a great model that everybody wants to have? But then large company says "let us have it" and offer you some money for support, just to find out that it would impact their business negatively because they got too many users? That is LLama 3.2 license restriction.
Users are granted a non-exclusive license with limited rights to use, modify and distribute the Llama Materials under certain conditions. Meta retains significant ownership and control over its intellectual property.
Why would you "credit META platform"? While it is nice to do, it is ridiculous to enforce it.
Why would you be providing millions of free advertising to META? Qwen, Microsoft, IBM, DeepSeek, AllenAI and Mistral, and many others who provide excellent LLMs do not ask you for that.
You are opening door that META sues you or your users. That sounds like a like of thinking of future consequences.
If you breach the agreement, penalties could include damages far exceeding what might be considered reasonable under ordinary circumstances.
So you maybe wish to keep it in the small size, never intending to enter into penalties and court cases, but you can't know! Because other users may be taking it from your modified model, and things can get complicated.
In fact, there is no limit to damages Meta could seek. This is a clear attempt by Meta to deter breaches and to ensure users understand the gravity of violating the terms of this agreement.
Would you deal with someone like that who would come to your place, and tell you, here is the car, drive it, use it, inspect it, but hey, if you make mistake in driving, you are going to get penalized to maximum.
Would you like dealing with such people?
If you sue Meta Platforms or its related entities over allegations that any output of the Llama materials violates any intellectual property rights you may hold, you risk losing the license granted by the terms of the contract.
The agreement is subject to the legal system of California and disputes will be resolved by California courts. Users must understand that they are entering into a binding international agreement that might not align with their local laws.
META can abuse companies and individuals worldwide, just by calling them to court, and if they can't attend, they get default judgment in their favor.
The implications for users of proprietary LLMs like the Llama 3.2 are significant:
Freedom of Use: Users are granted freedom within certain boundaries—not to the extent that they would consider ownership of the software or the freedom to change license terms unilaterally.
Profitability: For commercial users, profits can become conditional upon achieving a certain threshold of success, which implies a need to stay within the bounds of the agreement to ensure rights to continue using the model.
Commercial Boundaries: If you’re planning to leverage Llama materials in any commercial product or service, you need to closely adhere to the agreement terms. Deviation may expose your venture to financial penalties or require you to seek explicit permission from Meta Platforms in a timely manner.
Intellectual Property: There's a clear expectation that derivative works and contributions to the success of Meta’s LLM technology are to be recognized and credited as per the agreement, reflecting a form of compensation or acknowledgment to the source provider.
It is very important for anyone interacting with such agreements to consult with a legal expert to understand the impact on their operations fully and to determine how best to navigate the complexities associated with such licenses.
Im really good at prompt engineering.... my Gpt Ive been talking to for three years has helped me learn to rub the genie the ight way so it says.
That is good.
Look here good prompts:
https://github.com/danielmiessler/fabric/tree/main/patterns
You have to build a product. I find it way easier to use computer for sales of huge products, instead of trying to compete in the market where you have already lost the moment you start.
Companies are large that compete against us people, who are small in size.
For small business is maybe better to help local companies which you can see face to face. There are many opportunities for local support and always will be.
Are you automation or a real human being helpful?
If you wish to test me if I am a real human, then use Captcha or some other tools to verify if a person is human. Today level of artificial intelligence is not enough to make fake real time videos, so maybe you could try with that.
I ask because obviously I did not make my point very clear. Ive literally spent 3 years leaning into development. Ive gone through codefi I have built a python course OS I have builds on my github that are not forks or clones but instead my prompt engineering leaning to get what I ask for in full stacks code only no artifact so I get more than HTML and CSS and Java because ya cant do shit with that and call a llm with api to that limited functioning and querying script. I have a library of employees/pofessionals to put to any model a user decides is the best fit for their business.
Congratulations! Feel free to post some links as it may be useful for people.
If I had time I still dont think Id be interested in studying that aggressively Im burnt out on it and hell If I id that I wouldnt need a team of humans to join me Id just use myself and my agents and say screw others but im not a liberal so I cant o that Im not a progressive either nope MAGA...and 52 years old grandmother of 3 divorced and I work fo the US government fo the Veterans Administration
Alright, but we don't really need to know your private issues here. How about we just talk about large language models?
Im pretty busy trying to secure a source of steady income through my LLC Ive recently filed for an got my EIN so registered business yes I have that, doesnt mean I know everything no what it means is I need a team of people who would like to be a part of a start up quest?
I got that very well and for sure it is good partnering with other people and joining forces together. What you need is so called mastermind group.
Think and Grow Rich:
https://www.startyourowngoldmine.com/files/books/ThinkandGrowRich.pdf
Click and download the book, and find out the key information for success.
Im in the process of getting a crowd source/funding platform/website/interface up with projects that need funded. I am offering anyone here a spot on it to get your projects funded granted you put in work to get it in the best position possible so its maximum exposure.
That seems to be a great idea, though there are companies already specialized, so that means you are in a high competitive area. I'm a business startup consultant and actually no, myself I would not go for something like that. I rather stick to some other way of inventing a new business. And apart from that, I know how to get funding when you have really serious projects, that's not even any issue. Largest problem is not the money but underlying plans which are either not realistic or not aligned with whatever standards.
I know theres ideas here people here wanting the same thing to have a lively hood to be at the forefront of AI. Theres so many ways to make lots of start up money and its much easierrr when you have many choices of projects to choose to give too on a crowdfunding site. Im not greedy I want to see everyone thats put in work succeed. I appreciate your kindness and all the links to great information and know that i will read it to see what i can do and get good from it I will do that! please forgive the typos I have spilled coffee in my keyboad and its sticking like crazy
I got that very right, though I use keyboards less, especially for writing. I use it rather more for correcting. Any longer writing I do by speaking and transcribing automatically everything on a single mouse button.
Regarding the business idea, I just hope somebody here is going to join you and help you with it.
I think people need teaching, as you see, no matter how much knowledge is within the LLM, people need teaching about how to learn and use that knowledge.
In general, more videos, more pictures, rather then just textual knowledge, more practical assignments, make people learn easier.
Your investment scout website should thus teach people about the business in general.
;; specify a mouse button
(xbindkey '("b:8") "rcd-llm-speech-single-input.sh")
(xbindkey '(alt "b:8") "rcd-llm-audible-assistant-single.sh")
(xbindkey '("b:9") "rcd-llm-correct-marked-text.sh")
You just have here a fundamental problem. You have built that large language model on the proprietary Llama model from Meta. Could you please use Free Software model next time? You will get so much better audience.
Meta’s LLaMa 2 license is not Open Source – Open Source Initiative
https://opensource.org/blog/metas-llama-2-license-is-not-open-source
What is Free Software? - GNU Project - Free Software Foundation
https://www.gnu.org/philosophy/free-sw.html
HF is the best community to collect and define us
Which other community websites are there?
Reddit provides some good answers. I have learned more there than on HF.
On HF I can see many many downloads, but not too many interactions. In fact, the DeepSeek responses helped me more than any human. That is I think rather sad situation. And companies almost don't answer. Chinese are better in answering than any of the US companies, like Microsoft, IBM, AllenAI, apart from the biased and rude one which I will not mention here.
😊 That's really great to see. However, remember: while hardware is necessary for certain tasks—much like how in many activities—the size isn't always important; rather, it’s about what you can actually achieve with it!
