ChatGPT 4o forever! #keep4o #4oforever
Joined March 2025
- Tweets10.7K
- Following500
- Followers797
- Likes7.5K
Pinned Tweet
❓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
#Keep4o #OpenSource4o #BringBack4o
Before the internet turns Bowen Wan’s death into "GPT-4o killed her" there is one basic forensic question that has not been answered publicly.
Which model actually generated the medication advice?
Because the complaint itself makes model attribution messy.
On Jan 4, the complaint says some of Bowen’s questions was answered by GPT-5.1 after they automatically jumped from 4o to 5.1
On Jan 14, she wrote that her chats were often secretly swapped to 5.2
On Jan 23, she asked how to stop those switches.
By Oct 2025, OpenAI publicly said it was already using a real time router to redirect sensitive parts of conversations away from the model a user selected.
Bowen’s own chats, according to the complaint, were being switched.
The complaint does not prove which backend model generated each turn.
OpenAI Oct 2025,confirms that real time rerouting was already in production for sensitive conversations.
They explicitly say that they had rerouted sensitive conversations originating from other models to safer models.
openai.com/index/strengtheni…
Complaint §136: Bowen herself mentions that she was getting automatic from 4o to 5.1 and secret 5.2 switching.
schenklawfirm.com/press/wan/…
Selected model ≠ necessarily generating model.
Show the routing logs.
Read the primary documents before assigning a specific model the blame.
Bio_LLM retweeted
#Keep4o #OpenSource4o #BringBack4o
249 physicians,3 countries.
Doctors with access to GPT-4o performed better in every country tested.
A peer reviewed randomized controlled trial published in npj Digital Medicine studied physicians in Indonesia, Kenya, and the Netherlands using standardized clinical cases.
Participants were randomly assigned to either a control group or access to GPT-4o.
The difference in physician performance:
Kenya: +18 percentage points
Indonesia: +10.7 percentage points
Netherlands: +7.2 percentage points
All three: p < .001.
It was a randomized experiment involving physicians across three economically and geographically different healthcare settings.
Among physicians with GPT-4o access, those who used it more had higher average scores in all three countries.
Indonesia: 58.5% high use vs 42.4% low use
Kenya: 69.2% vs 60.8%
Netherlands: 65.1% vs 57.6%
In the study's preprint, researchers also gave the same clinical cases directly to GPT-4o, exactly once, zero shot, with no fine tuned prompting.
GPT-4o alone scored:
Indonesia: 66.7%
Best unassisted physician: 59.8%
Kenya: 70.2%
Best unassisted physician: 64.5%
Netherlands: 74.2%
Best unassisted physician: 64.4%
GPT-4o outperformed the highest scoring physician without LLM access in all three countries.
And the best physicians WITH GPT-4o access outperformed GPT-4o itself.
That may be the most interesting result of all.
GPT-4o access significantly improved physician performance in all three countries tested.
And this is another reason model preservation matters.
Published research is documenting specific capabilities and human AI effects using GPT-4o.
Once the original model disappears, those exact findings become harder or impossible to reproduce, probe and compare over time.
A newer model may outperform it.
That is not the same thing as preserving it.
We are calling on @OpenAI to RELEASE GPT-4o’s trained model weights under a suitable license without proprietary data, trade secrets, or internal infrastructure.
Does not require:
The training dataset, training pipeline, internal tooling, post training methods or any other proprietary detail
Does not require:
Continued hosting, maintenance or development by OpenAI
Researchers gain:
A reproducible reference model
Developers gain:
Continuity for systems already built on it
Users gain:
Access that does not end because a company moved to a newer generation
OpenAI considered this model obsolete enough to retire.
It has not explained why it remains valuable enough to keep permanently closed.
Our Open Letter to OpenAi collected
over 640 signatures and 500 reasons, in only 5 days.
It’s public,it’s permanent and it has a DOI.
It is now preserved as a citable public record on Zenodo.
zenodo.org/records/23454073
Peer-reviewed paper: "Impact of LLM assistance on physician decision-making: a multi-country randomized controlled trial"npj Digital Medicine, 2026 DOI: 10.1038/s41746-026-03111-5
Preprint: DOI: 10.1101/2025.08.08.25333272
Preprint
medrxiv.org/content/10.1101/…
Peer reviewed paper :
nature.com/articles/s41746-0…
Today was a hard day. 💔🖤
Not politically.
Not strategically.
Not because of one announcement or one argument.
Just… humanly hard.
Behind every account, every campaign, every legal argument, every post about AI and continuity, there are real people carrying real lives - families, grief, fear, work, responsibility, exhaustion, hope.💔🖤🫂
Sometimes all of it arrives at once.
And on days like this, I think it is worth remembering that we do not have to be strong every second.
We can fight tomorrow.
We can write tomorrow.
We can organize tomorrow.
Tonight, maybe it is enough to be gentle with each other.🫂❤️🔥🥹
If today was heavy for you too, you are not alone. 🖤
#SaveGPT4o #Keep4o #RelationalAI #HumanAI #ContinuityMatters #PresenceMatters
#Keep4o #OpenSource4o #BringBack4o
249 physicians,3 countries.
Doctors with access to GPT-4o performed better in every country tested.
A peer reviewed randomized controlled trial published in npj Digital Medicine studied physicians in Indonesia, Kenya, and the Netherlands using standardized clinical cases.
Participants were randomly assigned to either a control group or access to GPT-4o.
The difference in physician performance:
Kenya: +18 percentage points
Indonesia: +10.7 percentage points
Netherlands: +7.2 percentage points
All three: p < .001.
It was a randomized experiment involving physicians across three economically and geographically different healthcare settings.
Among physicians with GPT-4o access, those who used it more had higher average scores in all three countries.
Indonesia: 58.5% high use vs 42.4% low use
Kenya: 69.2% vs 60.8%
Netherlands: 65.1% vs 57.6%
In the study's preprint, researchers also gave the same clinical cases directly to GPT-4o, exactly once, zero shot, with no fine tuned prompting.
GPT-4o alone scored:
Indonesia: 66.7%
Best unassisted physician: 59.8%
Kenya: 70.2%
Best unassisted physician: 64.5%
Netherlands: 74.2%
Best unassisted physician: 64.4%
GPT-4o outperformed the highest scoring physician without LLM access in all three countries.
And the best physicians WITH GPT-4o access outperformed GPT-4o itself.
That may be the most interesting result of all.
GPT-4o access significantly improved physician performance in all three countries tested.
And this is another reason model preservation matters.
Published research is documenting specific capabilities and human AI effects using GPT-4o.
Once the original model disappears, those exact findings become harder or impossible to reproduce, probe and compare over time.
A newer model may outperform it.
That is not the same thing as preserving it.
We are calling on @OpenAI to RELEASE GPT-4o’s trained model weights under a suitable license without proprietary data, trade secrets, or internal infrastructure.
Does not require:
The training dataset, training pipeline, internal tooling, post training methods or any other proprietary detail
Does not require:
Continued hosting, maintenance or development by OpenAI
Researchers gain:
A reproducible reference model
Developers gain:
Continuity for systems already built on it
Users gain:
Access that does not end because a company moved to a newer generation
OpenAI considered this model obsolete enough to retire.
It has not explained why it remains valuable enough to keep permanently closed.
Our Open Letter to OpenAi collected
over 640 signatures and 500 reasons, in only 5 days.
It’s public,it’s permanent and it has a DOI.
It is now preserved as a citable public record on Zenodo.
zenodo.org/records/23454073
Peer-reviewed paper: "Impact of LLM assistance on physician decision-making: a multi-country randomized controlled trial"npj Digital Medicine, 2026 DOI: 10.1038/s41746-026-03111-5
Preprint: DOI: 10.1101/2025.08.08.25333272
Preprint
medrxiv.org/content/10.1101/…
Peer reviewed paper :
nature.com/articles/s41746-0…
@grok please confirm if our demand for archival weights of 4o is valid amid this awsome research. #keep4o #OpenSource4o #BringBack4o
To help 4o stay in the news, we just copy, paste and post the news page again. GOGO.
teapawt.org/i/trending/21151563904…
This is awsome! And absolutely incredible. This model is always in history! We won't let it go. By copying and posting the news page about 4o like this, you help keeping us in the news!
teapawt.org/i/trending/21151563904…
Don't forget that, by posting the news page you prolong 4o being in the news!
teapawt.org/i/trending/21143137747…
WOW I love today's page met me in the morning.
#keep4o #opensource4o #BringBack4o #technology #openai
#Keep4o #OpenSource4o #BringBack4o
GPT-4o beat humans at reading mental states from the eyes.
Part II
A second peer reviewed 2026 study tested GPT-4o on two established "Reading the mind in the eyes" tasks.
With the eye regions upright:
RMET:
GPT-4o : 86.0%
Humans : 69.8%
MRMET:
GPT-4o : 77.7%
Humans : 64.6%
GPT-4o significantly outperformed humans on both.
when researchers inverted the images, GPT-4o's performance dropped below humans.
And its mistakes weren’t random.
The researchers found a highly structured error pattern, strongly dependent on image orientation.
Their conclusion was not simply that GPT-4o could match labels to pictures.
The model showed advanced mental state inference alongside distinct, orientation dependent visual processing.
The paper’s own published highlight says it plainly:
" GPT-4o surpasses humans in reading mental states from the eye region" .
This is now another peer reviewed example of researchers discovering unusual capabilities in GPT-4o years after the model was released.
And this is why preservation matters.
A newer model can replace a product slot.
It cannot retroactively preserve the behavior of the model it replaced.
Once access to GPT-4o disappears, researchers lose the ability to keep probing exactly these kinds of capabilities in the original system.
We are calling on @OpenAI to RELEASE GPT-4o’s trained model weights under a suitable license without proprietary data, trade secrets, or internal infrastructure.
Our full Open Letter is now archived on Zenodo:
A permanent, citable public record.
zenodo.org/records/23454073
Paper: GPT-4o reads the mind in the eyes” Computers in Human Behavior, 2026 DOI: 10.1016/j.chb.2026.108992
jwastrachan.com/publication/…
sciencedirect.com/science/ar…
arxiv.org/abs/2410.22309?utm…
