GPT-5.6 Sol, along with Terra and Luna, will launch publicly this Thursday

131 pointsposted 3 hours ago
by jfrbfbreudh
(twitter.com)

63 Comments

aarvin_roshin

2 hours ago

Thoughts[^0] from Theo, who had early access:

> It's a damn good model. Not quite as "smart" as Fable, but it is incredibly capable. Fixed all the problems I had with GPT-5.5.

> It is incredibly determined. Will run for a day without even using a /goal. It understands subagents incredibly well and is great at orchestrating. It's super pleasant in use cases like OpenClaw and Hermes Agent. It knows iOS dev incredibly well.

> It has rough edges too, but FAR fewer than 5.5 did.

> For many things, gpt-5.6-sol will become my obvious defaults.

> It is better about [following instructions] than 5.5 was. Understands intent well and hammers until it gets there. Sometimes a bit too hard.

Also[^1]:

> gpt-5.6-sol is world leading in computer use. It made me use it 100x more. When we lost access to 5.6, I quickly started to go insane without it

[^0]: https://nitter.net/theo/status/2074708892341481755 [^1]: https://nitter.net/theo/status/2074720467395756499

bashtoni

an hour ago

I feel like listening to Theo about anything technical is like consulting a Labrador retriever for advice on quantum physics.

Every time I've ever seen one of his videos it's pretty clear he has very little understanding of development or engineering. I first became aware of him from his early "unit tests are a waste of time" stuff, and it seems his skillset is building a personal brand. Fair play, he's clearly talented at that, but that doesn't make his opinion on anything else worthwhile.

patates

32 minutes ago

> it's pretty clear he has very little understanding of development or engineering

I cannot prove it but I have a feeling that you may be conflating "he clearly has different opinions on things I consider non-negotiable" to "he doesn't know what he's talking about".

I also watched a lot of his videos. I wildly disagree with him a lot of times, but he has his reasoning, and I can see (and verify!) that those ideas are coming from an engineering perspective.

steve_adams_86

a minute ago

He's clearly very knowledgeable about some things, but I think he has harmed his credibility be becoming a 'tuber who prioritizes thumbnails and hot takes over engineering.

Havoc

26 minutes ago

And half his videos are him coming up with indirect ways of saying look how amazing I am.

ttoinou

40 minutes ago

There is a whole religion about tests that is worth attacking though

bashtoni

23 minutes ago

Sure. If his take was "100% unit test coverage is a waste of time" I think that's not unreasonable. You could make a case that the "you must write tests before you write code, every single time!" stuff is needlessly dogmatic. I also think that sometimes people focus too much on unit tests to the detriment of end to end tests that better model actual system interactions.

None of these were Theo's take. He was pushing the idea that unit tests in general were a waste of time because you could be shipping new features instead.

https://www.youtube.com/watch?v=pvBHyip4peo for an example of this. The nicest possible interpretation on this is that he's deliberately saying something he knows is wrong to attract attention.

jychang

an hour ago

> Not quite as "smart" as Fable, but it is incredibly capable.

THIS IS BECAUSE GPT-5.6 SOL IS... just a more posttrained version of GPT-5.5, not a brand new bigger model than GPT-5.5. It's not like how Mythos is bigger than Opus.

OpenAI switching to Sol/Terra/Luna renaming is just a way to rip off people and charge more money for the same sized model.

GPT-5.6 --------> GPT-5.6 Sol

GPT-5.6-mini ---> GPT-5.6 Terra

GPT-5.6-nano ---> GPT-5.6 Luna

Except OpenAI is about to charge way more money for GPT-5.6 Sol and GPT-5.6 Terra, than if they named it GPT-5.6 and GPT-5.6-mini.

deadlast2

22 minutes ago

I use Open AI and Claude a lot right like a lot everyday for hours multiple hours. Open AI gives much more value for money than Claude much more I'd say x 10. Mainly I use it for writing fiction books and literally Claude is locked 90% of everyday trying to jip me for tokens. It's not as good at coding for what I do which is a very complicated application. However it is very good at writing it's really good which is why I keep it right but over 90% maybe actually all of my work except the initial draft of a chapter is done by open AI.

threatripper

36 minutes ago

My feeling is that GPT-5.5 doesn't lack the raw intelligence so much as it lacks "methodology". I don't know how exactly to put it... how to approach a problem, how to take care of the details and side effects, how to handle unexpected difficulties and bugs, how to not spin out of control, how to write solid code, how to clean up afterwards, how to document, how to give useful feedback... the things that you learn on the job.

So, if they improved a lot in those areas, then GPT-5.6 could become a lot more useful compared to GPT-5.5 even though it might score lower in many benchmarks. It's possible but unlikely since their approach was mostly brute force in the past.

ppaattrriicckk

39 minutes ago

> OpenAI switching to Sol/Terra/Luna renaming is just a way to rip off people and charge more money for the same sized model.

Excuse me, but what are you on about?

Unless I'm mistaken, they have literally(1) stated that it will cost $5 per 1M tokens in, and $30 for 1M output tokens. The same as GPT-5.5.

[1] https://openai.com/index/previewing-gpt-5-6-sol/

behnamoh

19 minutes ago

OpenAI already has a Mythos level model, it's called GPTCyber and before that, it was called gpt-5.5-pro.

isoprophlex

2 hours ago

Increased tenacity & goal following is exactly what I want in this model, to make it compete with Claude models.

(A little toning down of the goblin fetish would be nice too, haha.)

stingraycharles

an hour ago

“Understands intent well and hammers until it gets there. “

If there’s anything I learned over the past 12-18 months is that this is a recipe for disaster, except for throwaway stuff.

I thought most senior engineers settled on the fact that steering a model yields much better results?

icepush

42 minutes ago

It's very possible that would be the best strategy over the last 12-18 months and now that this is released it is no longer the best strategy.

stingraycharles

8 minutes ago

That would be an extremely massive leap if agents could suddenly make nuanced architectural decisions and prevent technical debt.

In my experience even Fable still requires guidance (although the options it provides are generally better).

oefrha

44 minutes ago

I wouldn’t call it a recipe for disaster, but oh boy if you leave an agent that “hammers until it gets there” on its own with an underlying bug in a dependency…

laurels-marts

2 hours ago

Damn this is exciting. I love that gpt models are much faster, efficient and cheaper than Claude models. They are so fast even on high/xhigh that I don’t find myself using the parallel agent setup anymore much since its cognitively less demanding to just follow along what the model is doing and most tasks it will complete in <5-<10mins anyway.

jychang

an hour ago

THIS IS BECAUSE GPT-5.6 SOL IS... just a more posttrained version of GPT-5.5, not a bigger model than GPT-5.5. It's not like how Mythos is bigger than Opus.

OpenAI switching to the Sol/Terra/Luna naming is just a way to rip off people and charge more money for the same sized model.

GPT-5.6 --------> GPT-5.6 Sol

GPT-5.6-mini ---> GPT-5.6 Terra

GPT-5.6-nano ---> GPT-5.6 Luna

Except OpenAI is about to charge way more money for GPT-5.6 Sol and GPT-5.6 Terra, than if they named it GPT-5.6 and GPT-5.6-mini.

Two important things to note, if you want to verify what I say/correct me:

GPT-5.6 Terra actually scores worse than GPT-5.5 on many benchmarks. It's not GPT-5.5 trained with more compute; it's basically GPT-5.6-mini that's been distilled from GPT-5.6 full size. Remember, GPT-5.4-mini had almost the same benchmarks as GPT-5.2 after all.

Opus 4.8 runs at ~90 tokens per second. Fable 5 runs at ~40 tokens per second on from Anthropic, because it's a bigger/slower model. A few days after the release, when the dust dies down, look at how many tokens/second GPT-5.6 Sol is running at. I will bet it's the about same as GPT-5.5, and not half the speed. (OpenAI is not incentivized to slow down the model for paying customers). But the model tokens/sec will be a big clue- if OpenAI is charging more money for the same sized model or not.

sk4rekr0w

43 minutes ago

bigger doesn't mean better, chill out

reassess_blind

2 hours ago

I’m bouncing back between Codex and Claude like a ping-pong ball. I much prefer the experience using Codex, less verbose and to-the-point I’ve found. But Fable, being as strong as it is, is a big draw for Claude right now. I’ll likely switch back to Codex if 5.6 Sol is comparable.

avereveard

an hour ago

Same. For some reason late opus model are very superficial doing ux work and so am using gpt for that, but backend is much better engineered by claude, gpt prefer to duplicate everything it needs on the spot causing class sprawl

bredren

35 minutes ago

How are y'all carrying context history from one agent to the other?

I also flip between the models due to quota, TUI enhancements, model updates and service availability.

To handle this, I built a thing that normalizes your transcripts between Claude Code and Codex into a shared DB, then a CLI and skill.

It has made it so it doesn't matter what I built where (or when) I just refer to the work and drop in a /total-recall (or $total-recall on codex) and the agent brings it into the current convo.

I realize there are a lot of ~memory tools out there, but I think particular my approach and product behavior is unique.

If you're open to giving it a try, I'd appreciate any feedback: https://contextify.sh recent show hn: https://news.ycombinator.com/item?id=48777790

tyteen4a03

32 minutes ago

Why not both? T3 Code exists.

reassess_blind

21 minutes ago

Hadn't heard about it. Launching either "codex" or "claude" from cli isn't exactly a pain point I was looking to fix.

Edit: I guess the real answer is that I don't want multiple subscriptions. I use one for a month, and then decide if I switch to the other.

behnamoh

18 minutes ago

How? I'm pretty much locked into Claude Code and even if gpt models are good now, the experience with codex CLI has been so bad I won't go back to it.

e.g., it still doesn't have /revise or /undo!

bottlepalm

22 minutes ago

Coding with AI it feels like if you're not using the best model then you're possibly missing out - creating less capable, maintainable, just plain 'good' code. Why waste time using anything less than the best and cleaning up the mess later on. This is why I feel like local models and Chinese models aren't taking off (and Gemini/Grok) - they work, but they're plain just not as good as OpenAI/Anthropic. If you have the money then it doesn't make sense to code with anything else.

armchairhacker

a few seconds ago

Does Fable write better code or just can solve more problems?

TheCoreh

17 minutes ago

There are diminishing returns, especially for more mundane tasks. Fable is nice, and I bet Sol is also nice. But there really isn't much of a difference right now when using something beyond Opus or presumably Terra for most things. They're most useful when doing greenfield, highly complex/novel tasks. When Open Source catches up, it will be more widely adopted.

Philip-J-Fry

16 minutes ago

That depends entirely on how you're using AI. If you're getting it to do all the hard thinking, then sure using the best model is probably always going to be better. But it's also going to be expensive.

Using cheaper models and using your skills and expertise from the pre-AI era can get you working just as fast. You've gotta be more specific about the work you need doing. It's less "vibes" based, but they're still effective.

Also, Chinese models absolutely are taking off. I used Claude and GPT at work, and then I tried using some Chinese models for personal projects. I am 100% convinced they're like 90% as good for 10% of the cost. But you've basically gotta be a good developer first and know what you want and know when it's giving you shit.

mhrmsn

an hour ago

Is this the reason Anthropic extended use of Fable 5 via subscriptions until July 12? Seems a bit like it

matheusmoreira

an hour ago

Probably, but I think it's too little too late. Not much point to it if it's not permanent. The "get the most out of Fable until it goes away" frenzy is getting old fast. The cybersecurity blocks are very obnoxious too.

If OpenAI can launch a Fable tier model that's actually usable on a subscription, then Anthropic is just going to lose, and badly.

matheusmoreira

an hour ago

Will it be available on subscription tiers? That will get me to switch away from Anthropic.

ray__

2 hours ago

Any previewers have hot takes? I've really preferred gpt-5.5 over Opus 4.8 for data analysis and scientific software work. It seems much more reliable. Fable is unusable for the type of work that I do (due to guardrails). Really looking forward to trying these new OpenAI models out.

CjHuber

2 hours ago

Interesting to hear people like gpt-5.5. For me it feels smart only at one shot prompts, but if you try to build up session context before doing something it feels magnitudes inferior to Claude. I'm almost sure its because the thinking of previous turns is stripped with the responses API, so if I tell it to analyse something deeply, what remains of the understanding in future turns is only the short response text of that analysis

porker

2 hours ago

For gpt-5.5 I build up that session context into a markdown file, and then I start a new session and give it the markdown file with the instructions for what to do.

I'm guessing this works better because it can always go back and re-analyze the saved context.

prodigycorp

an hour ago

Do you have a source for this? I'm pretty sure responses api is only there to obfuscate reasoning, but that they're still keeping reasoning traces in the backend.

tedsanders

23 minutes ago

I work at OpenAI and can confirm that's correct: reasoning tokens are discarded after each new user turn (though not after each message or tool call).

Our docs show a diagram here:

https://developers.openai.com/api/docs/guides/reasoning

> Input and output tokens from each step are carried over, while reasoning tokens are discarded.

Keeping reasoning tokens around is better for caching and for remembering past insights, so you might reasonably wonder why we designed it this way. The main benefit of dropping reasoning tokens is that you can fit a lot more work inside the model's context window before you're forced into a slow and lossy compaction step. This was a larger consideration with our earlier reasoning models that had shorter context windows (~200k), longer thinking times (up to ~100k per message), and poor compaction. However, now that we've shipped longer context windows, we've trained our models think much more efficiently, and we've made compaction way better than it used to be, the balance of factors is changing. Tune in Thursday!

wahnfrieden

an hour ago

There's no source for it as it's made-up nonsense

throwuxiytayq

an hour ago

> thinking of previous turns is stripped with the responses API

Why do they store an encrypted reasoning payload in the session file and pass it to the API? Just a ruse? Reasoning isn’t even that many tokens, you think they’d degrade their model quality like that?

Reasoning messages would be lost immediately after a single tool call, unless you mean they sometimes go back and strip the reasoning channel retroactively, but that would increase costs via cache invalidation. I just don’t see any way it would make sense for them to do.

And wouldn’t this be noticeable by reasoning tokens not being accounted for in the context window usage?

pavpanchekha

2 hours ago

For compiler work I found that Sol is noticably better than 5.5 (and I generally use OAI models because I like the Codex app), but Fable was still obviously better.

ottoboney

2 hours ago

It seems comparable to Fable to me in my uses.

scottmf

2 hours ago

What types of use cases?

wahnfrieden

2 hours ago

That's great to hear - and for the same price as 5.5, and reportedly with much lower token use per task.

Were you able to try Sol Ultra?

ottoboney

2 hours ago

No, my organization limits access to xhigh.

petesergeant

an hour ago

Interesting, data analysis work is the only thing I’ll use Gemini for

aarvin_roshin

2 hours ago

I'm most curious about whether OpenAI finally taught its models how to design interfaces. They have been behind the other labs in this area for what feels like ages.

MrBuddyCasino

2 hours ago

What do you feel is the best model for interface design right now?

Dibes

an hour ago

For me claude is the best, hands down. Fable took it a step even further.

simianwords

17 minutes ago

Earlier I predicted that Fable and Sol would be of similar capability, I think I will be wrong. Here is why: there is no indication that there are any classifiers like in Fable. I think OpenAI found out how to lobotomise the model without classifiers but the tradeoff is that it is a weaker model. I wonder how people feel about that. Would you like a highly intelligent jagged model with classifiers or slightly less intelligent smooth model without classifiers?

simianwords

an hour ago

I find codex way more usable. It’s not pretentiously verbose like Claude. It’s also responsive - I can see the progress easily and steer the conversation. With Claude, it might take 15 minutes and I would lose patience.

gck1

8 minutes ago

Both are verbose in their own way, and both - terrible. Claude models love to throw huge blobs of text in architecture planning / interview conversations, but in not a mentally draining language. OpenAI models are more compact, but very dense & formal - they will speak in RFC language for a button that clicks and submits a form.

So claude: 10 paragraphs of prose

codex: 1 paragraph of jargon over jargon.

lillecarl

an hour ago

I held out on OpenAI until last month because I despise Sam Altman, but using Codex is a great experience and 5.5 (medium) I'm on 20$ is very capable, follows instructions when it should and confronts me/challenges me when it should.

UX is nicer where the agent is somehow "separated" from execution.

yzydserd

an hour ago

The question is, launch to who …

Tiberium

an hour ago

To everyone.

steve-atx-7600

40 minutes ago

"We’re expanding preview access globally now." Preview access? Not as straightforward as "launching on thursday".

tedsanders

3 minutes ago

I believe what the post meant to communicate is:

- alpha testers will start getting access now

- everyone will get access Thursday (barring banned countries / individuals)

Historically, some companies and individuals have gotten alpha access before public launches, to give feedback and adapt their products to the new models. With GPT-5.6, some folks had alpha access, but this was taken away while the model was being evaluated and approved. Now, alpha access will be enabled for some partner companies in the next two days before our wider launch.

I work at OpenAI.

SkiFire13

37 minutes ago

Honestly they sound like pokemon game names.

minimaxir

an hour ago

...so is it a good idea to use up all my Codex quota by Thursday in the hopes of a reset to promote GPT 5.6?