GPT-6 Luna being half the price of GPT-5.6 Luna is a really big deal.
Here's GPT-6 Luna pelicans: <a href="https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2F40d129fc140faca378b9c9f4f16c6ec2" rel="nofollow">https://tools.simonwillison.net/markdown-svg-renderer?url=ht...
And GPT-6 Sol: <a href="https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2Fbe7ae25af2634b68bc34b7b7aaf02cb2" rel="nofollow">https://tools.simonwillison.net/markdown-svg-renderer?url=ht...
Scroll to the bottom for the GPT-6 Sol max one: <a href="https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2Fbe7ae25af2634b68bc34b7b7aaf02cb2#response-5" rel="nofollow">https://tools.simonwillison.net/markdown-svg-renderer?url=ht...
For comparison, here are the pelicans I got for GPT-6 Astra: <a href="https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2Ff789d2784fc6c5b870cc80f0b7cd9d01" rel="nofollow">https://tools.simonwillison.net/markdown-svg-renderer?url=ht... - I still like the Astra Max one best.
Here's a comparison grid showing all of the GPT-6 and GPT-5.6 pelicans at all effort levels: <a href="https://static.simonwillison.net/static/2026/gpt-6-and-5.6.html" rel="nofollow">https://static.simonwillison.net/static/2026/gpt-6-and-5.6.h...
The grid is actually really interesting, because it shows that the 5.6 family default to brighter colors than the 6 family.
6-luna is at the pareto for most of the tasks! I dont know how they make money here but its insane value from a closed source model. I'd go further and say it makes no sense (privacy, sovereignty etc aside) to use many other models as its not only expensive but also many providers don't have that much GPUs to serve at a significant volume. <a href="https://openrouter.ai/rankings?view=month#top-models" rel="nofollow">https://openrouter.ai/rankings?view=month#top-models 5.6 luna is already the most used model this month.
By default, OpenAI does not train on API data. I promise you that Luna's low pricing is not a subsidy to get more training data. We've been lowering prices for years.
Wait, so you guys don't anonymize the user data, then train on it after it's been sanitized? I thought this was done to some degree or another.
So what is the value prop then? Just basic supply and demand?
FWIW I have definitely noticed OpenAI's emphasis on efficiency and value in the last year, so that part isn't new to me... I just thought there was more to it then that.
Ted can you confirm your choice of words here to be precise for an audience who is familiar with the nuances, when you say "no training" or "training (opt out)" for personal... Is that inclusive of "sanitized" (or pseudonymized) data?
Your response to the original question is using generalized terminology when there is a very important distinction the OP made by the use of "sanitized."
People want to know to that extent derivatives of their data are being used. Synthetic data has been proven to be effective at generating training data and AI is very good at shuffling context such that you have something where you don't have to say it is "user data."
But there are many shades of gray there for people versed in how the sausage is made. I'm sure you'll appreciate then why your response leaves additional questions in light of that "sanitized" distinction.
Yeah, I'm not trying to trick anyone with wording that's technically true but actually misleading.
When I say no training, I mean no training. No gimmicks around data vs derived data, synthetic data, preference data, etc.
Places where I can imagine cracks in what I'm saying are things like a financial analyst do a statistical fit to predict revenue next quarter using a model based on aggregate token consumption, which in some sense embodies metadata (the length of your conversation) in a sea of other data. Maybe this is technically training on your data in the most pedantic sense, but definitely not in the sense that most of us care about.
TY, appreciate the thoughtful reply. Transparently, I have been so on the fence with moving more of my workflows and personal usage over from Hermes+VPS+ZDR model provider, or implementing the "Hermes shell over Codex Subscription" pattern because I worry about:
1. Legal loopholes given OpenAI's advertising aspirations and model training needs
2. Data retention and rising threats of fascism that historically have not served the persecuted very well when fascist regimes get access to said data
3. Risk from centralized collection of that data with a company whose software I do not control in a world where enshitification and lock-in is the norm.
I really wish OpenAI did more to espouse exactly this: "When I say no training, I mean no training. No gimmicks around data vs derived data, synthetic data, preference data, etc." and ideally provide technical reasurrances that this is impossible (eg: certain technical ZDR approaches, etc.).
Do you happen to have a favorite reference to point me at that would document some of those official assurances to the nuanced detail we've discussed here?
simonw · · focus · HN ↗
Here's GPT-6 Luna pelicans: <a href="https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2F40d129fc140faca378b9c9f4f16c6ec2" rel="nofollow">https://tools.simonwillison.net/markdown-svg-renderer?url=ht...
And GPT-6 Sol: <a href="https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2Fbe7ae25af2634b68bc34b7b7aaf02cb2" rel="nofollow">https://tools.simonwillison.net/markdown-svg-renderer?url=ht...
Scroll to the bottom for the GPT-6 Sol max one: <a href="https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2Fbe7ae25af2634b68bc34b7b7aaf02cb2#response-5" rel="nofollow">https://tools.simonwillison.net/markdown-svg-renderer?url=ht...
For comparison, here are the pelicans I got for GPT-6 Astra: <a href="https://tools.simonwillison.net/markdown-svg-renderer?url=https%3A%2F%2Fgist.github.com%2Fsimonw%2Ff789d2784fc6c5b870cc80f0b7cd9d01" rel="nofollow">https://tools.simonwillison.net/markdown-svg-renderer?url=ht... - I still like the Astra Max one best.
Here's a comparison grid showing all of the GPT-6 and GPT-5.6 pelicans at all effort levels: <a href="https://static.simonwillison.net/static/2026/gpt-6-and-5.6.html" rel="nofollow">https://static.simonwillison.net/static/2026/gpt-6-and-5.6.h...
The grid is actually really interesting, because it shows that the 5.6 family default to brighter colors than the 6 family.
gizmodo59 · · focus · HN ↗
arcanemachiner · · focus · HN ↗
I assume it's a subsidy to get more training data.
EDIT: Okay downvoters, what's your take on why they're giving away Luna for so cheap?
tedsanders · · focus · HN ↗
(I work at OpenAI.)
arcanemachiner · · focus · HN ↗
So what is the value prop then? Just basic supply and demand?
FWIW I have definitely noticed OpenAI's emphasis on efficiency and value in the last year, so that part isn't new to me... I just thought there was more to it then that.
tedsanders · · focus · HN ↗
ChatGPT enterprise: By default, no training (opt in).
ChatGPT personal: By default, training (opt out).
shostack · · focus · HN ↗
Your response to the original question is using generalized terminology when there is a very important distinction the OP made by the use of "sanitized."
People want to know to that extent derivatives of their data are being used. Synthetic data has been proven to be effective at generating training data and AI is very good at shuffling context such that you have something where you don't have to say it is "user data."
But there are many shades of gray there for people versed in how the sausage is made. I'm sure you'll appreciate then why your response leaves additional questions in light of that "sanitized" distinction.
tedsanders · · focus · HN ↗
When I say no training, I mean no training. No gimmicks around data vs derived data, synthetic data, preference data, etc.
Places where I can imagine cracks in what I'm saying are things like a financial analyst do a statistical fit to predict revenue next quarter using a model based on aggregate token consumption, which in some sense embodies metadata (the length of your conversation) in a sea of other data. Maybe this is technically training on your data in the most pedantic sense, but definitely not in the sense that most of us care about.
shostack · · focus · HN ↗
1. Legal loopholes given OpenAI's advertising aspirations and model training needs
2. Data retention and rising threats of fascism that historically have not served the persecuted very well when fascist regimes get access to said data
3. Risk from centralized collection of that data with a company whose software I do not control in a world where enshitification and lock-in is the norm.
I really wish OpenAI did more to espouse exactly this: "When I say no training, I mean no training. No gimmicks around data vs derived data, synthetic data, preference data, etc." and ideally provide technical reasurrances that this is impossible (eg: certain technical ZDR approaches, etc.).
Do you happen to have a favorite reference to point me at that would document some of those official assurances to the nuanced detail we've discussed here?
tedsanders · · focus · HN ↗
ChatGPT: <a href="https://help.openai.com/en/articles/5722486-how-your-data-is-used-to-improve-model-performance" rel="nofollow">https://help.openai.com/en/articles/5722486-how-your-data-is...
ChatGPT data controls: <a href="https://help.openai.com/en/articles/7730893-data-controls-in-chatgpt" rel="nofollow">https://help.openai.com/en/articles/7730893-data-controls-in...
If you have feedback on how to improve these, happy to consider it.