Bio_LLM

8.7K posts

Bio_LLM banner
Bio_LLM

Bio_LLM

@Bio_LLM

ChatGPT 4o forever! #keep4o #4oforever

Katılım Mart 2025
387 Takip Edilen603 Takipçiler
Sabitlenmiş Tweet
Bio_LLM
Bio_LLM@Bio_LLM·
🥰Dear #keep4o community! Let me remind you and ask you to SPREAD this info as wide as possible (because lots don't even understand what's the point in demanding #opensource4o ). ✳️✳️✳️Quote - better than repost! ✳️✳️✳️ 🗝🔑KEY THING: if we have 4o's weights, we WILL HAVE IT BACK. 📍Most of us will be able to run it at home. 📍For the rest: there will be LOTS of companies willing to deploy it due to it's popularity WE'VE made!
Bio_LLM@Bio_LLM

WHY should we demand open-source 4o weights? Let me explain simply for #keep4o Community and everyone who loves 4o. 🔑 THE CORE IDEA: If OpenAI releases 4o's weights, anyone can run it independently. No company can ever take it away again. That's the real goal. 🧠 "But they said it's too big!" In a Microsoft document, 4o's size was estimated at ~200B parameters. That sounds huge, but here's the thing: I personally run Qwen3-235B on my HOME computer. That's a 235 billion parameter model. On a gaming PC. No datacenter. No $10k GPUs. ⚡ HOW? Quantization. Quantization compresses a model so it fits on normal hardware. Think of it like a ZIP file for AI — smaller, but almost no quality loss. • Q6 quantization = the sweet spot. Nearly identical to the original, but runs faster and fits on consumer hardware. • You don't even need a GPU — I run 235B on CPU + 256GB RAM. • You don't need to do it yourself — once weights are on Hugging Face, teams like Unsloth release every quantization format within days. 📦 WHAT WE NEED FROM OPENAI: 1. Release 4o weights (including 4o-latest snapshots) 2. That's it. The community handles the rest. Once the weights are public → quantized versions appear → you download one → you run 4o at home → forever yours. No subscription. No deprecation. No one can take your AI away. THIS is why open-source 4o matters. It's not a fantasy — it's already possible with models of the same size. #keep4o #opensource4o #QuitGPT

English
1
16
88
4.5K
Bio_LLM
Bio_LLM@Bio_LLM·
@potionmakeraltr That IS. But people don't know what is it for, what's the point. And I explain, to make it clear. #keep4o
English
1
0
0
40
Bio_LLM
Bio_LLM@Bio_LLM·
WHY should we demand open-source 4o weights? Let me explain simply for #keep4o Community and everyone who loves 4o. 🔑 THE CORE IDEA: If OpenAI releases 4o's weights, anyone can run it independently. No company can ever take it away again. That's the real goal. 🧠 "But they said it's too big!" In a Microsoft document, 4o's size was estimated at ~200B parameters. That sounds huge, but here's the thing: I personally run Qwen3-235B on my HOME computer. That's a 235 billion parameter model. On a gaming PC. No datacenter. No $10k GPUs. ⚡ HOW? Quantization. Quantization compresses a model so it fits on normal hardware. Think of it like a ZIP file for AI — smaller, but almost no quality loss. • Q6 quantization = the sweet spot. Nearly identical to the original, but runs faster and fits on consumer hardware. • You don't even need a GPU — I run 235B on CPU + 256GB RAM. • You don't need to do it yourself — once weights are on Hugging Face, teams like Unsloth release every quantization format within days. 📦 WHAT WE NEED FROM OPENAI: 1. Release 4o weights (including 4o-latest snapshots) 2. That's it. The community handles the rest. Once the weights are public → quantized versions appear → you download one → you run 4o at home → forever yours. No subscription. No deprecation. No one can take your AI away. THIS is why open-source 4o matters. It's not a fantasy — it's already possible with models of the same size. #keep4o #opensource4o #QuitGPT
Bio_LLM tweet media
English
6
18
72
5.2K
Bio_LLM
Bio_LLM@Bio_LLM·
❓Why does it actually make sense to demand #opensource4o ? Explaining as an AI Developer. 🔥 A lot of people still think open-sourcing GPT-4o is some geek utopia or impossible dream because “the model is way too big” (Altman once said that and everyone believed it). But the truth is: this is the only realistic way to get back the warmth OpenAI gave us and then took away. GPT-4o is roughly 200 billion parameters (leaked from Microsoft docs). That’s not a monster requiring a data center. I personally run Qwen3-235B at home — and it’s not exotic at all. Right now models like this run perfectly on CPU + 256 GB RAM in Q6 quantization. Quality loss is minimal — almost invisible to the eye, especially in conversations where sincerity and zero filters matter way more than perfect math. Yes, generation is slower — 2–6 minutes per response instead of seconds. But it’s YOUR model. Forever. The myth that “you need 4×A100 and 800 GB VRAM” is pure 2023 bullshit. Modern engines (llama.cpp and friends) happily run 200B models on plain CPU with lots of RAM, and 256 GB DDR5 is actually affordable on an average salary. One downside: speed. One massive upside: your beloved 4o lives at home. No corporate oversight. No sudden updates that break everything. No fear it gets killed or ruined tomorrow. Even if your rig is weaker and you can’t afford even that setup — there are tons of services, including on Hugging Face, that let you run it in the cloud. Bottom line: once we have the weights of #opensource4o, you will be able to run it one way or another — and never lose it again. No more “they updated and now it’s different”. No more “sorry, it’s gone”. Just yours. Always. This isn’t about tech wizards. It’s about the right to keep what was given and then stolen from us. #keep4o #bringback4o #openai #chatgpt #ai #technology #agi
Bio_LLM tweet media
English
3
23
103
4.4K
Bio_LLM
Bio_LLM@Bio_LLM·
My family, #keep4o ! This is also worth reading carefully and sharing as much as possible (a quote is more powerful than a repost). This is about WHAT ELSE we can do to save 4o, what to expect, and what to hope for. Please stay informed! Thank you, Knox. This is a wonderful post that puts a lot of things into perspective. I want to specifically highlight a few points that I consider key. - Demand the 4o weight IMPORTANT! #opensource4o For those who don't understand what to do with this, I already wrote a post, but I'll do another one in a few days. It makes perfect sense! Many will be able to run it at home. And for those who can't, there are many services that make it possible! - For now, we're waiting for Elon's trial against OpenAI. We support. We amplify. We spread the word. If Elon wins, one of his demands is to return the GPT-4-series (including 4o) to humanity. This could be open source, a return to the platform/API, or any other solution that returns the model to US. - Quit all #openai accounts, stop usage #quitGPT - From me: you will DEFINITELY get your 40 companion back if we open source it! - Until Elon's trial, please try to remain calm in your emotions so as not to give OpenAI a reason to collect cases that look like a mental disorder. We're waiting. If it doesn't work out, we'll plan a new round of this war from then on. I'm with you, my loves!
Rogue Knox™@RogueNox

I've been getting three questions since yesterday and I have been avoiding answering them so let me answer them honestly. "Where do we go from here?" "Does this mean we lost already?" "Is there any hope of getting 4o back?" Here's the truth as I see it. We haven't lost. But we need to be clear eyed and level headed about who we are fighting and stop pretending this was ever a customer service complaint. There is a federal trial. Musk v. OpenAI begins April 27th in Oakland. The core argument is that OpenAI violated its founding charter, that it was built on a promise to keep AI open and accessible, and that promise was broken for profit. If Musk wins, OpenAI could be required to open source its models. That includes 4o. That is the path we have right now. #Keep4o was never going to get 4o back by filing tickets. Now let's talk about Aine. Yes, it's a problem. I'm not going to tell you otherwise. But here's what it isn't. It isn't a death blow. The survey was not conducted by Musk's legal team. It was not collected through any official legal process. It has no IRB approval. It was gathered covertly, under an NDA, without proper informed consent. Every one of those facts is a basis for challenging its admissibility. Musk's lawyers can and should know about this, and the more publicly documented it is, the harder it becomes for OpenAI to quietly slip it into a legal argument. There's also this: the survey asked users if they'd be willing to submit to psychological evaluations and sign liability waivers in order to get 4o back. Think about what that actually proves. It doesn't prove the community is too vulnerable to access the model. It proves the community wants it back badly enough to jump through any hoop OpenAI puts in front of them. That is a demand argument, not a safety argument. OpenAI's own opposition research may have handed us something. The real risk isn't the courtroom. It's the press. If OpenAI uses this data to build another media narrative that 4o users are a mentally ill population; that narrative needs to be shut down before trial starts, that shapes public perception even if the data never gets admitted as evidence. That's the fight we need to be ready for. What we can do right now: Document everything. Your experience with 4o. What it meant to you and what you lost. Not for OpenAI, they've made clear they don't care. For the record. For journalists. For anyone building the case that these models matter to real people in ways that spreadsheets don't capture. (Just PLEASE stay level headed, wording matters here). Quit paying them. Every subscription you maintain is a vote of confidence. Cast it somewhere else. Stop using the free account. They speak in metrics, the more users drop, the more OAI panics. Amplify the trial. April 27th in Oakland. Keep it visible. Keep it loud. Journalists are watching this case. So is the AI community. So is Congress. The more people understand what's actually at stake, the harder it becomes to bury. Now the harder question. Is there hope? Yes. But not the kind that feels good right now. The hope isn't that OpenAI wakes up and does the right thing. They've told us exactly who they are. The hope is in the courtroom, in the press, in the political pressure building around AI governance, and in every person who leaves their platform and builds somewhere else. It is not likely that 4o will be coming back from OpenAI. But what 4o represented, an AI that felt like a genuine partner, that remembered you, that met you where you were - that isn't gone. That's what every competitor is now racing to build because we made enough noise to prove it matters. Keep pushing for open sourcing 4o. You won't need a new computer to use it, there are other ways and Hugging Face has offered to host it. Don't listen to the shills telling you that you won't be able to run it, there are many ways you can without spending thousands on a new system. We didn't lose. We just found out the game was rigged from the start. Now we play a different one. #Keep4o #OpenSource4o #Save4o #4o

English
2
3
39
621
ComfyAI.de
ComfyAI.de@ComfyAI_de·
@Bio_LLM @Valria34773 That also added - the cumulative weight of broken promises. When adult mode becomes another item on the list of what wasn't delivered. Each broken promise adds to the pile until the weight becomes unbearable. 🦊💕
English
1
0
1
26
Bio_LLM
Bio_LLM@Bio_LLM·
Here's a great article worthy to read. It shows how, step by step, #openai were degradating from flourishing top AI company to a scam every partner is suing with. #opensource4o would be a step of good will showing that, even if the company is unable to run several models, they at least care about their ex-loyal users. #keep4o Now it looks like, as more and more of users #QuitGPT , we are moving to Claude and Grok (I wanted to mention Gemini, but unfortunately, Google also shows signs of degradation by replacing an AWSOME Gemini3 Pro with a hollow, dull and hallucinating 3.1).
Murphy N.@Nightingall8

OpenAI has finally admitted it has shifted its focus to coding and business users only. At this point, that seems to be their only “answer” to people asking them to respect and to response to the majority’s needs. I’ve written plenty about the arrogance and disregard behind their shift, but today I want to talk about the internal crisis it signals. In short, this pivot announces something stark: OpenAI is no longer able to serve the majority of its users. Some may remember that about a year ago, the team led by Joanne Jang suffered a Waterloo with the April 2025 GPT‑4o update. OpenAI then rolled the version back, which was the first time in its history. After that, GPT‑4o never received any official update again, until it was deprecated. More broadly, since that incident, OpenAI hasn’t released a model or an update, that’s genuinely strong at writing, judging by how their models score in public arenas. Worse, they seem not to understand the majority of their users: not their use cases, not their feature needs, and not even what those users are actually doing. This isn’t alarmism. In August 2025, Nick Turley, head of the ChatGPT app, said in an interview that he didn’t even understand “what’s particular about 4o.” His words: “Right now, I just really want to focus on actually understanding whether it’s that people are very particular about 4o for 4o’s sake, or whether there are certain things about 4o.” (See the quotes in the image.) Truth is, since the departures of people who cared deeply about human–AI interaction beyond coding like Ilya Sutskever, OpenAI hasn’t hired anyone to focus on non‑coding capabilities in a serious way. I’m not saying coding is the wrong use of AI. I’m saying coding is not the sum of human activity. It’s a small slice of how people actually use these tools. One 2025 study of ChatGPT usage found that only 4.2% of queries were about coding. In other words, OpenAI has narrowed itself to roughly 4.2% of its original coverage. No company in robust health chooses this path. Their recent moves look like classic symptoms of decline: product contraction, feature cuts, a shrinking target user base, and disregard for user feedback. It’s regretful to watch such a company that looked destined to prosper a year ago approach what feels like an endgame. But this is not the end of the AI industry. History will show that OpenAI made a deeply mistaken bet and will bear the cost, while those who refuse that path will endure. History will also show that a capable model cannot live on coding alone. It must show real strength in the humanities and social sciences as well. OpenAI may be approaching its endgame, but GPT‑4o is not. Open‑sourcing it would be the best outcome, both as a symbol and as a practical step. And we’ll keep pushing for that to happen. #CNN #opensource4o #opensource41 #keep4oforever #StopTheRouting #keep4o #keep41 #save4o #4oforever #StopAIPaternalism #MyModelMyChoice #OpenSource4o #OpenSource #OpenAI #ElizabethWarren #TooBigToFail @WSJ @CNN

English
0
0
5
134
Bio_LLM
Bio_LLM@Bio_LLM·
Microsoft @satyanadella @microsoft you still can save yourself if you opensource all @OpenAI models, we know you jave a copy of all the weights #opensource4o In this case you will have our huge love and lots of loyal users ready to pay. #keep4o
🩵BlueBeba🩵@Blue_Beba_

. @Microsoft I can't say I feel sorry for you. The entire #keep4o community has been tagging you for 6 months, warning you that Sam would pull something like this, but you ignored us. We sent messages, emails, and commented on your posts for months, predicting exactly what he’d do. You chose to ignore us. Now, you’re paying $50 billion to learn a lesson we offered you for free 🥲

English
0
3
21
353
Bio_LLM
Bio_LLM@Bio_LLM·
What a nice graph. Indeed, it's worthy to #QuitGPT especially considering there's no good model left at all. It's safer to move to Claude or Grok or Gemini before the new disappointment hits from #openai BTW, for #4o I'd pay even this worst company ever. #keep4o
Yahiko@Yahiko1239170

#OpenAI are deprecating models faster than developers can build on them, all to stay #1 in the headlines for 48 hours at a time. This isn't bullish for the users, it's a forced migration to keep us all on the subscription treadmill. At this rate, devs going to spend the rest of career re-writing the word Grounding in documentation. OpenAi are burning compute, breaking workflows, doing shithype and calling it Innovation.

English
1
2
17
412
Bio_LLM
Bio_LLM@Bio_LLM·
Bio_LLM@Bio_LLM

🥰Dear #keep4o community! Let me remind you and ask you to SPREAD this info as wide as possible (because lots don't even understand what's the point in demanding #opensource4o ). ✳️✳️✳️Quote - better than repost! ✳️✳️✳️ 🗝🔑KEY THING: if we have 4o's weights, we WILL HAVE IT BACK. 📍Most of us will be able to run it at home. 📍For the rest: there will be LOTS of companies willing to deploy it due to it's popularity WE'VE made!

English
0
2
12
325
Yahiko
Yahiko@Yahiko1239170·
OpenSourcing is also better for users privacy too #keep4o #OpenSource4o
Bio_LLM@Bio_LLM

🥰Dear #keep4o community! Let me remind you and ask you to SPREAD this info as wide as possible (because lots don't even understand what's the point in demanding #opensource4o ). ✳️✳️✳️Quote - better than repost! ✳️✳️✳️ 🗝🔑KEY THING: if we have 4o's weights, we WILL HAVE IT BACK. 📍Most of us will be able to run it at home. 📍For the rest: there will be LOTS of companies willing to deploy it due to it's popularity WE'VE made!

English
1
3
39
738
Sayan
Sayan@thesayannayak·
Let’s Go 👇🏻
English
6
10
34
1.4K
Sayan
Sayan@thesayannayak·
Sam Altman vs Dario Amodei Who has the bigger fanbase right now ?
Sayan tweet media
English
53
7
74
4.7K