Serving cheaply at that scale means routing, batching, and cache hits, not a better model :D
Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users
51–60 of 281 posts
Re: Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users
#52They must really be feeling the commoditization pressure. I'm not sure what the way out of this is, ChatGPT and Claude are still good products, but they are not necessarily premium products anymore. I expect a few things to happen in the next year: 1) Exclusive MCP server deals/API integrations 2) Significant switch to B2B marketing, even moreso than we've seen before, with API interfaces being paid and chat-client i…
I'm not sure I agree 1) Back then, even as a free user you'd be able to use the strongest model (even if with tight limits). Now, you need to pay to use Sol, and you need to pay to use Opus or Fable. It does seem fairly premium in that sense. Idk about 5.6 Luna, but the previous Instant was really bad, even for very casual users. It would hallucinate non stop. 2) When $100 and $200 per month plans launched, they were…
Back when coding for me still meant copy-paste from the web version, it was only worth the $20/month for me.
They only added the $100 Pro plan in April during GPT 5.4 times.
Today I happily pay $400/month for Codex and Claude Code.
Re: Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users
#53Nothing in the ChatGPT model release notes yet: https://help.openai.com/en/articles/9624314-model-release-no... (This is a subtle nudge at anyone from OpenAI who reads this to make sure they get updated.) OpenAI have a model called "chat-latest" - I wonder if that's running this new model yet: https://developers.openai.com/api/docs/models/chat-latest It's described as "points to the latest Instant model currently use…
There’s no 5.6 instant I think — 5.6 Luna is gonna be the new instant tier model
https://gist.github.com/simonw/aae4febd3c6f7bc5b7811857edb3c... has screenshots that still show "Instant" as an option for ChatGPT Chat... but not for ChatGPT Work.
Re: Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users
#54Re: Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users
#55Earlier quoted context omitted.
There’s no 5.6 instant I think — 5.6 Luna is gonna be the new instant tier model
No, Luna is the new free model. https://gist.github.com/simonw/aae4febd3c6f7bc5b7811857edb3c... has screenshots that still show "Instant" as an option for ChatGPT Chat... but not for ChatGPT Work.
It’s hard to understand this .. like sure we can select instant but is there an actual model called 5.6 instant? Like is it on LM Arena and OpenRouter or available via API etc
5.5 instant is definitely A Thing it’s even name checked in this OAI post
Re: Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users
#56Giving free ChatGPT users access to reasoning (the 'Think' toggle) will have a broader impact on the world than every new paid model and coding agent combined.
Because everyone now outsources much of their thinking and researching to LLM's, our collective culture + brain is shaped in a cyclical manner by using them.
It's the mechanical homogenization of culture and groupthink.
Re: Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users
#57Its always fun to try to read between the lines here to speculate why they are doing this. Maybe Luna efficiency gain was actually significant enough that putting all the free users and giving them super generous limits makes sense. They might be doing this to improve the messaging of AI among causal users since right now there is a huge amount of datacenter backlash in the US due to AI grievances. Maybe they have to…
Re: Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users
#58At first I thought the Sol updates was perhaps trying to help with some complaints of Sol burning through tokens, complaints that have prompted some new data points on https://codex-resets.com/ . But seeing the graphic with the visual weather report: that makes me think that is not the goal at all. :)
What a bizarre example of a more direct answer. Any human would simply say “No, it doesn’t rain here in summer.” Even after identifying the 0% chance of rain, it still drags the conversation on and on and on
Re: Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users
#59Giving free ChatGPT users access to reasoning (the 'Think' toggle) will have a broader impact on the world than every new paid model and coding agent combined.
Going from 5.5 Instant (which was noticeably bad) to 5.6 Luna is a big jump as well. OpenAI is probably the most prominent among the general public - an advantage in some respects but they're giving away a lot of free inference and thus have to use a pretty small model to do it.
Re: Improving GPT‑5.6 Sol in ChatGPT, expanding GPT‑5.6 Luna access for free users
#60Earlier quoted context omitted.
It's also pretty cheap if you're paying for it (for coding). But I don't quite trust it for anything complicated, so I often use Terra.
Terra is awful and not cheap enough to make up for it. I'd suggest using luna or sol.