GPT-5.6
openai.com
openai.com
> Intent understanding: GPT-5.6 can better infer the user’s underlying goal and intended level of work without you specifying every step. Continue to state important constraints, approval boundaries, and success criteria explicitly.
> Original image detail: GPT-5.6 preserves the original dimensions of images sent with original or auto detail instead of resizing them to a patch budget or pixel-dimension limit.
> Use shorter prompts: In internal evaluations, replacing long, explicit system prompts with minimal prompts improved scores by roughly 10–15%, while reducing total tokens by 41–66% and cost by 33–67%.
> Avoid generic brevity instructions: GPT-5.6 is more sensitive than GPT-5.5 to instructions such as “Be concise,” “Keep it short,” or “Use minimal text.”
> Control warmth: GPT-5.6 does not become meaningfully better when prompted to be broadly friendlier or more empathetic.
That part is confusing because it's not like they provide an example of how default GPT-5.6 output compares with GPT-5.5 both with default output and prompted for brevity. Whenever I use such prompts, it's usually because I want the model to give me the gist in a few sentences. I'd be stunned if GPT-5.6 was that concise by default. I would think that could "break" a lot of things for developers who didn't know to make prompt changes after upgrading to 5.6. What if you were expecting GPT to be as wordy as it usually is? Then suddenly your output is not wordy enough?
Smells like OpenAI trying its best to stave off financial armageddon for another few months. Then again, I'm not sure why they chose to waste so much output computation on verbal diarrhea all this time up to now.
Remains to be seen how the "shorter prompts" advice translates to homogeneity/collapse though.
This is a trap.
It's the optimistic fallacy that poisons all "consumer scale" machine learning products and what's going to effectively ruin these models as they keep chasing it in the same way that web queries were ruined, social media feeds were ruined, and media recommenders were ruined.
For the vendor, optimizing metrics across their whole user base, they always see positive technological progress as their system gets better at making assumptions and accumulating user engagement scores in aggregate. But for the individual user, most of which has some weird tail intent/interest and some of whom have many weird tail intent/interests, the experience quietly but catastrophically degrades. Output/results become more generic, more divergent with the underspecified "weird tail" intent, and more stubbornly hard to ever wrangle towards that "weird tail" altogether.
We've been watching this cycle happen for 20 years now and it's proving hard for anybody to escape because it works so well for the trillion dollar company driving it forward. But while each step might feel ergonomic and welcome to individual users, there's a frog boiling enshitification at play.
In pursuit of output quality and capability (rather than simply the vendor's user count), what we need rather than "makes better guesses" is "presses for more clarity", even where it feels kind of annoying.
Even among human professionals, one of the first hurdles of breaking out of junior tier work is gaining the confidence to press your colleagues and clients to be more specific in their thoughts and expressions despite their desire to have you do it all for them. But they're often coming to you with incomplete, muddy, and conflicting ideas for which there is no safe and correct assumption that you might just run with, and it's your expertise (i.e. relevant "intelligence") that's critical to bringing attention to that. To achieve professional progression, you need to learn to do that and to not just optimize appeasing the ambiguous client/colleague today in exchange for mutual expense tomorrow. To avoid enshitification, which is probably not possible, we need these models to be learning that too.
> GPT-5.6 does not become meaningfully better when prompted to be broadly friendlier or more empathetic. Instead of generic instructions such as “Be friendly and warm,” use concrete guidance: > Be direct and tactful. Acknowledge friction specifically when relevant. Avoid canned reassurance and unnecessary sign-offs.
Soo basically, my new 5.6 custom instructions: Be Jeeves and eliminate all friction from my life through immense processing power. Acknowledge friction specifically when relevant. Avoid canned reassurance and unnecessary sign-offs.
[1] https://developers.openai.com/api/docs/guides/latest-model#c...
What about my favorite, "no yapping"?
This will totally make it brain damaged over a certain tasks. Sort of like the same brain damage that prompted OpenAI project managers to destroy ChatGPT.app today.
> Avoid generic brevity instructions: GPT-5.6 is more sensitive than GPT-5.5 to instructions such as “Be concise,” “Keep it short,” or “Use minimal text.”
I don't follow. Isn't "the model actually cares and will do what you say" a reason to use those kinds of instructions more liberally?
I used to go to a barber and if you said "cut it short", he cut it really short.
(For that matter at what point is it "long"? And does the rest of the context matter? Should it be short too?)
Does this mean ChatGPT will stop botsplaining things to me? I get it quite a bit more per unit time from ChatGPT than claude. Maybe that will change now.
(By botsplaining I mean when the AI explains some unstated premise of the prompt itself back at me as a correction when in many cases it's the motivation for the question in the first place)
y'know, I don't think I will. I really, truly want one-word answers to any binary or multiple-choice question. If I want more, I will ask for it once the model has given its answer.
RIP Caveman skill. Six month good. Now skill dead.
A shorter prompt results in half as much tokens spend? I find this very hard to believe.
When has this ever not been the case? I don't think this is a GPT 5.6 specialty!
> Intent understanding: GPT-5.6 can better infer the user’s underlying goal and intended level of work without you specifying every step. Continue to state important constraints, approval boundaries, and success criteria explicitly.
Winner by default!
Pathetic situation, this one, where we are supposedly building a superintelligence while at the same time thinking that fasting is a biological weapon.
Seems a bit more hand picked than usual to me..
Sol is the first verified frontier model to ever beat an ARC-AGI-3 game
Fable's main advantage is that its average solution size is smaller. However, GPT 5.6 Sol is a substantial improvement from GPT 5.4/5.5 which would write verbose, defensive code. 31KB for GPT 5.4/5.5 down to 26KB for GPT 5.6 Sol, with better performance for Sol.
Fable scores slightly lower, but with an average solution size of 12.2 KB.
Replay of Sol attempting the game: https://arcprize.org/replay/83543d22-8e1e-439a-8809-129ff1d9...
It seems a weird and arbitrary challenge for a language model to be expected to perform. It also seems like there are some harness/visual issues even in the first few steps, where it states that it hasn't moved when it clearly has.
Yikes
Also what does this tell about Yann LeCuns whole world model theory? Bro has been going on and on about it. He has made multiple wrong predictions on the trajectory of LLMs.
At some point his claim should be fully falsified no?
What's the consensus today on codex vs claude code, does it really matter anymore?
One thing I appreciate with Codex is, OpenAI nowadays sometimes just gives you quota resets you can bank, so when you use up weekly quota before the week ends, you could just reset the quota, to continue using Codex. I've been much less anxious about Codex quota because of this perk. I just used one reset in the bank yesterday, and still have 3 resets left. Whereas with Claude, when you've used 95% quota 3 days before the week ends, you'd be much more anxious.
On the other hand, Claude Code's /remote-control mechanism is extremely helpful when I am running it in the cloud and wants to monitor it or control it on my phone. Codex currently doesn't support this kind of usage. Codex only allows you to use your phone to connect to a session on your desktop, not in the cloud.
I personally find GPT-5.5 to be a better programmer than Opus 4.8, it is extremely thorough, but I don't like the code it generates ("austere"), and find Opus 4.8 to write more "human friendly" code. The programming comments GPT-5.5 makes is pretty awful where-as Opus 4.8 is good. I feel like Opus 4.8 is better at grasping my intention than GPT-5.5, and honestly find GPT-5.5 to be kind of "autistic". I do prefer the language (not the writing) of GPT-5.5, as I find the philosophical flowery language of Opus 4.8 kind of annoying.
I have only managed to try Fable 5 a little bit, which feels like a much more generally smarter version of Opus 4.8, that is much better a programming and grasping your intention, and I think even the intention of your code, and is _really_ good at spotting bugs or problems with logic in your code. It feels wicked smart but is extemely expensive. It feels smart in the sense like it has a "bigger brain" and is much more sensitive to subtleties/details.
These are different "brains", have different "personalities", etc. I think the best thing is to develop a feeling for it yourself.
After 6+ months of exclusive Claude Code usage, I was begrudgingly forced to try Codex once Anthropic rejiggered their limits such that I kept maxing out my $200/mo plan in just a few days. These days I pay both $200/mo plans, and it's just about enough to get me through a week's work (small game studio - infinite code to write!)
It's more diligent and empirical and results focused, and less creative. It sometimes needs a kick to avoid a Zeno's paradox of incremental steps to get to the goal. But it produces more reliable code with fewer race conditions, unhandled negative cases, etc.
It's also better value from a $$ POV, or at least has been. This fluctuates a bit.
You're also free to use your Codex subscription with other harnesses, like opencode, etc. Unlike Anthropic. Plays better with others.
Codex historically will follow tasks more closely with less creativity, whereas Opus will do more than you specify. I wouldnt consider either one better due to this fact, just makes them useful for different situations. Generally they'll perform similarly for most tasks.
Opus and Fable dominate 5.5 in artistic design (pixel art, ascii art), and edge out 5.5 slightly in general UI design taste. Have not tested Sol in that regard yet.
So far in my usage Sol has been superior to Fable at graphics rendering engine optimization.
Codex will work longer, and in single sessions without as much subagent usage.
Codex only has 256k context but its compaction is absolutely next level. You will not notice compactions and they will happen multiple times during a complex task or set of tasks without you ever having to notice or care. Claude code on the other hand still has fairly poor compaction.
Codex has more generous usage limits, and they also give you usage resets (weekly+5h resets) that you can bank for a month or so. Not sure how often they give these out.
Codex also seemingly never has outages or weird delays like Claude code does.
OpenAI randomly resets usage just like Anthropic does
I would use both if you code often
But what I love about Openai is that they still let you hook OTHER harnesses up to a subscription. My Pi setup has been built up for a few months now into exactly what I want and moving over to CC or even Codex is really annoying.
Caveat: I vibe code in tiny little chunks. I see what I want to do, and exactly how I want it done, then prompt that, refine, what was output, then repeat. I bet Fable is better at building a whole app from a 2-sentence prompt; but that's just not important to me at all.
They're different models with different philosophies behind them. This is anecdotal with a user group of 1, but in my experience:
Claude has a stronger personality and is more creative. If you give it vague instructions, it's better at filling in the blanks with reasonable ideas.
GPT-5.5 is better at following instructions. If you know exactly what you want, it will do it without going off the rails. It's also less likely to imply that you're dumb, but I don't really care about that. Some people do.
I had put a decent amount of effort into setting up that initial codex attempt and it went so poorly that i've been entirely uninterested in trying again. This was maybe a month or so ago, and i know stuff moves fast, but for me, i like the models, dont care for the harness.
For personal stuff, I've been pretty happy with chatgpt's $20 plan. I believe it has considerably higher limits than claude's $20 plan, and it's enough for the personal stuff I play with (hermes, and some small coding stuff). Also allows me to keep up to date on openai models.
Claude lost my trust around February this year when the plan would say nonsensical things as "delete this method" that was clearly a key method on that part of the codebase.
For personal projects I am using Codex 20$ plan and when that is over I use DeepSeek which is insanely good for the cost.
Codex is more details focused, often catches wonky bugs and correctness issues that Fable misses, feels more terse and less "friendly", more like a stern senior engineer versus a friendly talkative engineer (Claude). Codex is also better if you're already an engineer, Claude is better for non-engineers. I.e. Codex works better if you know exactly what you want and know the right way of explaining it.
Consensus is probably the wrong word for the popular opinions reflected in HN that you might get.
I would recommend that you have 2 of each at all times when it comes to AI so you don't necessarily become overly locked to quirks of one thing. You'll soon realize that things move so fast that you just start internalizing common patterns instead of depending on one specific vendor.
I recommend that you try pi and codex besides claude, to get your own feel for it.
- codex UI is much more responsive
- i get feedback about the progress easily
- the tool calls and results are very legible, I can click them and see the progress
- no one talks about this but the tool call and response notification are handled much more elegantly in Codex. In Claude Code, it is handled in a clunky way using loops which always causes some delay
- you can steer the conversation midway in Codex
- /side is underrated (/btw is the equivalent and is much worse in Claude Code)
- I have to admit subagents are handled better in Claude Code
I'm trying Codex as my primary the last day or so, because I'm at 98% use and reset in 3 days on Claude. I'm worried about a lot of our skills and CLAUDE.mds and the like getting lost unless I migrate them, but otherwise codex seems to be working great.
You get much more generous usage from the 20x plan.
And you get far better uptime.
If benchmarks and early tester impressions are accurate, you also get access to Fable level capability at greater speed and lower cost (included in subscription).
I have a bunch of Claude Code Plugins and yesterday asked Codex to make them accessible to itself. It wanted to rewrite most of it. I was hoping i could get by with some symlinks or something to avoid drift.
I have my context managed in a structured (but nowadays way too big) Obsidian Vault. I also built myself a vector based "vault search" capability and have my harness use this as a tool to find thematically similar things across the different contexts, when needed. I also build a few custom skills and extensions for my harness to be able to do my work.
Talking about harness: I use pi.dev and have taken care of, that i set it up in a way as to easily be able to switch the intelligence layer without loosing context. Yes, there are differences in how well models perform, but if a model refuses a task - like gpt-5.5 not willing to build a downloading tool for Annas Archive - I switch the model to something less finicky.
Thus I was able to switch to gpt based models after about a year with Claude (and having had a Claude Max since the early days it was available).
I played a lot with other models recently, to see how stabl my setup is for switching, should something like Fable happen on a broader scale with the US government. As said, minor changes in tonality, minor issues ith the quality of long text being written by the model, but most of it is actually managed in by the tonality docs, guard rails, coding standards and the likes, I set up over the last 9+ months of intensive work with it (first in Claude Code, then Codex and now as said pi.dev).
So YMMV and it heavily depends on your setup. But I more and more treat those models as interchangable.
I do almost all my regular coding tasks with Codex 5.5 on medium. Sometimes for niche edge cases, or when I run out of tokens on my Codex sub, I'll switch to Claude. Some recent examples where Claude was able to solve things Codex couldn't:
- 3D gamedev layout: I asked Codex to render a solar system in a certain camera positioning, saying it needed to fit the planets of the system to the viewport. Codex just couldn't do it, even on high reasoning: Claude Opus did it first attempt.
- Tricky Tiptap image drag-n-drop layout implementation: Codex failed this after numerous iterations. Claude Opus also struggled mightily to get it to work, but I think around 3 attempts it nailed it. Both of them ended up grepping the Tiptap code from node_modules - that's the kind of task it was.
But these are really isolated examples. Across all my projects (I have many; mostly TypeScript, but also things like C#), Codex "Just Works" (tm), with minimal prompting effort from me.
This is using the same AGENTS.md prompts, which were designed firstly for Claude use, so maybe it's something that could be optimized better if I understood gpt as well?
IMO the two biggest problems not really being answered by both OpenAI and Anthropic are: 1. Why not make specific models good at specific tasks for Codex/Claude Code. Theres a handful of types of work here whereby small good quality models would do better than these generalised all purpose models whereby someone discovers Fable is bad at biology.... 2. Why cant they consistently run these models and keep them performing? Performance of the models seems to directly correlate with amount of compute available, but they dont talk about it...
Personally, I find it very interchangeable. I open codex --yolo or claude with whatever there yolo flag is (have an alias).
I installed the Claude Code Codex skill provided by Anthropic and I am having Claude invoke it automatically to review all plans and changes. The nice thing about this is that for an additional $20/month pro plan I can extend the runway for Claude rate limiting and compare frontier model responses. I am looking for more ways now to work in Codex as a subagent that gets used automatically from Claude Code.
https://arena.ai/leaderboard/agent
5.6 isn’t on there yet but Fable leads by a significant margin atm
Claude's very bloated and convoluted by comparison. Maybe you need the bloat (Claude Design), but I prefer the more razor's edge efficiency of Codex.
Model wise, I can't really tell. They all do what I want them to do most of the time and go off the rails occasionally. The question is increasingly becoming who's faster and cheaper and gives me more tokens, not who's better.
You're fully free to use and try anything and without caring about what others think is right
Codex writes all of the code, no exceptions.
Works great, especially when you ask Claude to break up large CRs into roughly 10 minutes of Codex work each.
I didn’t think I could have found a better solution, spawning multiple subagents with different models is such a great thing.
I built in the past very small cli wrappers to call other models; Claude Code often refuses to do that, lies and does the job itself instead of delegating to another provider’s llms.
Codex with GPT 5.5 is much better at general SWE tasks but Claude Code with Opus is far better at complex reasoning tasks like reading and summarizing research papers, replicating experiments, identifying research gaps and proposing interesting follow ups.
Not having to deal with Anthropics constantly changing policies, token-gating, and carrot-and-stick marketing helps me to focus on work, rather than dealing with their company problems.
Between the two the biggest difference by far is ... getting your harness / AGENTS.md / skills / tools set up right.
- built-in image generation using your subscription, which can be super handy
- can actually edit Google Docs and Google Sheets (Claude can only create new or sometimes append)
- I get a surprising amount of mileage out of the $20 plan
They both have their places for sure.
Try Pi: https://pi.dev/
Codex and Claude Code are not mutually exclusive, you can use both.
I tried them both side by side, mostly for reviewing existing Godot/GDScript code, or sometimes generating Swift Mac apps, including converting ancient relics I wrote eons ago in Visual Basic on Windows
Codex was consistently better than Claude: https://i.imgur.com/jYawPDY.png
Besides the useless "This is good" findings while reviewing and the excessive "oops you're right" backtracking, Claude's atrocious UX and borderline "spyware" make me never want to try an Anthropic product again for a long long while.
Can they all be wrong/paid-off?
Comparing this to other models, I find it similar to GPT-5.5 and a bit behind Sonnet 5. You can see how other models fared here: https://senko.net/vibecode-bench/ (you can also fetch the prompt and the the 5.6 Terra resulting code on from that page).
I don't have access to Sol yet (on a Plus sub, which should get it according to what I've read), so can't do the more interesting test. I'll update the above page as soon as I get access - hopefully soon.
Which model is the best at the moment, for this kind of stuff, in your experience?
The naming convention is especially difficult to decipher depending on what your native language is. Of course a latin language speaker might be able to easily determine oh yeah each one is slightly bigger than the other but I still think it borderlines too confusing.
That aside all the numbers look amazing, and I'll be happy to probably main this alongside grok-4.5 for a while comparing the two on price and efficiency.
I vastly prefer the direction that OpenAI seems to be going with token efficiency and performance compared to Anthropic who seems to be moving towards a world where you just token-max as much as possible ignoring any and all costs.
I'd like to know how cherry-picked this is, and what tests it performed less overwhelmingly in, but I suppose that info is not going to be on this post.
If it pans out to be as good as it says, that's great. On the other hand, if this model is not overwhelmingly impressive over Fable, I will lose what remaining trust I had in these announcements.
By the way, this isn't about their 5.6 version in particular I guess, it's just the first time I've looked at one of their videos.
The eval is an agent that runs a set of tools and a prompt we can tune separately for different models. The OpenAI version of the prompt was specifically tuned based on their guide[0]. Then we let Opus to run another agent that acts as a user, trying to solve a problem (anonymized and taken from production). The problem is complex and we don't expect it to be solved by these agents, but we measure how the agents operate when faced with a vague problem:
- Opus 4.8 and GLM 5.2 both identified a constraint sooner and stopped so the user can fix an issue first that the agent cannot solve.
- Sol tried hard to solve the issue with different tools, burning tokens, until finally reached to the same conclusion with Opus and GLM. It was two times more expensive compared to Opus and six times more expensive to GLM for this task.
- Terra went even further and started calling tools that would not solve the issue, burning tokens and failing.
- Luna repeated the same failing tool call until it hit the round limit, and burned more money than Opus.
I'm kind of puzzled with the new GPT. Like, yes Sol is OK for programming, but I was expecting to get a cheap agentic model for non-programming tasks, one that can detect if things go awry and correct. Terra is too expensive and Luna not really fit for the task. Sonnet 5 is a bit better but more expensive than Opus 4.8, which is still the best in my evals. GLM 5.2 is extremely good if you can define the task and the tools clearly for it, and costs pennies!
[0] https://developers.openai.com/api/docs/guides/latest-model
And 5.6 Luna ($0.21) is also impressive, cheaper than GLM 5.2 ($0.37) with higher intelligence.
Here's my benchmark results for GPT-5.6:
https://aibenchy.com/?q=gpt-5.6
(the high reasoning variants are still running, uploading them soon too)
EDIT: The high variants are there too, enjoy the hamsters[0].
This one is really promising, as it may allow to close major gap with Claude in design/UI skills
Or if you want to see some in 3D, OpenAI featured a pelican riding a tricycle, bicycle, pony and another pelican in their livestream this morning: https://www.youtube.com/live/Wq45rvPGNHs?t=1070s
Amusing that they use A100e as the reference point to sound impressive. Different ways you could make that conversion, but based on FP4 FLOPs (yes it's disadvantageous to A100, that's the point), that's something like 200hr on a GB300 NVL72 rack.
Not nothing either, but far less astounding sounding than 700k hrs.
I don't know exactly what part of my codebase is triggering it, so I'm going to have to keep poking, but apparently the guardrails are not that gentle despite the phrasing. :(
One major sticking criteria for not going with OpenCode / pi for all of my coding is I want access to the tier-1 frontier model of the day without API pricing - e.g. afaik I can't use Fable 5 via pi harness even though I have a subscription, so for this week I'm on Claude Code. It's not the need to Fable 5 for everything, but even if I just want the marginal intelligence benefit to stress test an architecture decision, it's a safety blanket to know there isn't a ~smarter~ model I could have used. And for my use cases, the doggedness and capability of these frontier models has been insanely effective.
My feeling is we're still in the Uber era subsidy period - the moment the subscriptions either try to lock me in longer than a month or stop OAI/Anthropic stop delivering frontier models in the subscriptions, I'm out - switching fully over to pi.dev or another OS harness and routing my token spend via OpenRouter or offloading to Qwen locally. Then I'll have to put an accurate dollar amount on frontier intelligence.
They also seem to really not care about alignment, or care about it in the wrong way. It's entirely missing in the blogpost and there are some concerning bits in the model card, seemingly treating CoT controllability as something to be "investigated" rather than the warning sign it's supposed to be.
My question to previewers: how are the guardrails for random joe that wasn’t personally blessed by the ai pope to access the non-nerfed model? Fable is a nightmare in this regard, but I’m not sure whether 5.6 also gets a critical side-eye from the gubmint when you ask it to fix bugs in your code (you filthy hacker, you).
I noticed that Fable uses shell tools almost exclusively (even to search and edit files), compared to previous Anthropic models.
Having run some experiments with 5.6, I notice that it uses built-in file systems and provider native tools much more (not shell tools), compared to previous OAI models.
Some pretty big claims and results! Excited to see how it feels during usage.
I use Fable and 5.5 extensively and I still find both have a place in my toolkit, i.e. Fable IS good but it isn't perfect, and it's still better to play them off against each other. I have Fable and 5.5 write plans and have them adversarially review each other's plans.
Having this amount of competition in the coding model space is good for all of us.
GPT 5.5 has a tendency to write English calques and non-idiomatic prose in other languages. Although that can be somewhat tamed with detailed instructions and a corpus of confusing terms, the model’s output often reads like a literal translation rather than native prose. Since I notice these issues most clearly in languages I know well, it makes me reluctant to trust the model’s output in languages in which I’m less proficient.
Ironically, ChatGPT began as a simple text-generation tool, but much of its offerings and benchmarks now focus on coding and agentic workflows, while leaving behind what made it notable in the first place.
"This request requires additional safety checks, which can take extra time. Hang tight or retry with a faster model for a quicker response, though it may be less capable of handling complex requests."
At least it gave me the option of waiting instead of just unceremoniously downgrading me. Appears to be making progress but... weird?
Assuming I take the 5x plan it would give me about an hour of active sessions with terra ultra (maybe ultra is not good value regarding tokens?), not even using Sol yet. Does everyone using codex use the 200$ plan?
I normally use the 100$ anthropic plan and barely ever reach the usage limit.
I look forward to seeing how it compares once I have access. Not getting tripped by spurious safe guard flags could be an advantage.
I generally prefer OpenAI. I usually start projects with Codex; I love the plan we created. Then, after about 2 hours of working, back and forth, etc.I realize it's drifting HARD, or getting stuck on relatively simple things.
Once I get frustrated enough, I sometimes start over with Anthropic. Anthropic has been better (for me at least) to work with from start to finish.
I have the $200/month plan for each of them. (And I used to have the $250/month plan for Gemini... lol?)
Both Fable and Sol are good and definite improvements. I don't have a favorite yet. It's too early to tell.
The biggest difference I'm noticing is usage.
Anthropic is like a used car salesman. Not allowed to do a basic check on your own car (website), they'll scrape every penny possible out of you. Low limits, high api prices, trying to keep everything in their system.
OpenAI is like a cool but aloof dad. Go ahead and borrow his car, shoot, he'll even pay for your gas every once in a while. He'll answer your 5th question drunk, forgetting what you asked. But, at least he's a nice drinker that buys the group shots.
I have to watch my Fable usage, and I'm sure as hell not going to pay API prices. I don't have to watch/worry about my Sol usage even in Ultra.
But, 1 thing I'm noticing using Sol Ultra (on fast mode) is that it's slowing WAY down after a bit. I work the opposite of peak times so that's not it.
It has a tendency to do things without asking, a trait I'd associated more with the Claude Opus & Sonnet models than with Codex & GPT in the past. Specifically I've seen it go and update e.g. README.md files filling it with recenty-biased gibberish that means nothing to the user (e.g. very specific technical notes related to what it was currently working on) or staging and adding design/spec documents that were meant to just be working documents. In general it tends to behave more aggressively with git, if you let it get its hands on it. It has stronger "opinions" on that stuff, that don't always agree with me.
I'm going to have to update my prompts, I think. But I'm not used to this kind of thing in Codex, which in the past has been much more explicit and cautious, and one of the reasons I've preferred it over Claude.
It is very "smart." It also has a tendency to yak-shave things. Producing huge volumes of correctness and regression tests and nitting over e.g. very minor variances.
One thing that is "entertaining" is letting two separate instances review each other's code. They will endlessly find things to nit at.
It also defaults to 'low' mode for some reason. Can't tell if that's a step backwards compared to GPT-5.5 in medium mode so I'm sticking to medium.
Edit: just noticed it's spawning subagents in 'high' thinking mode.
UPDATE: it is now available in chatGPT account also, they rolled it out
The base model is certainly cheaper and more token efficient etc, but on large tasks cost in some way is now n^2
this seems very interesting
Parameter: reasoning_effort
Function tools with reasoning_effort are not supported for gpt-5.6-sol in /v1/chat/completions.
To use function tools, use /v1/responses or set reasoning_effort to 'none'.'
I hope we aren't trying to push customers off the chat completion endpoint... Responses endpoint looks great on paper, but the business wants more visibility and control over the reasoning process than this product currently offers.
Edit: This is broken in my VS copilot setup too.
E.g. for GeneBench Pro, it looks like you would always use GPT-5.6 Sol over Terra/Luna, its pareto optimal.
For Agents Last Exam, you would maybe want Luna, then Terra, then Luna, then Sol as you increasingly budget for tasks.
I feel that there may need to be a new auto mode in many of these cases. It selects the best model and thinking given a particular problem.
Feels like it's going to have to go that way eventually, because here we have about 20 different model and thinking levels you could use, and they're not obvious which ones are right for the given use case.
it's impossible to _try_ it out on release!
it's not on their codex subscription, or the web/mobile chatgpt interfaces, or aws bedrock, etc. I just cant find a working endpoint with the latest model after they announce
Fable had issues with the sourcing and organizing images, and shoot itself at foot looking for shortcuts as usual. As I was getting it fix these back and forth, I copied my prompt and gave it to Sol.
Sol has surpassed my expectations by far. With a one shot simple prompt on a complex task, it gave me a working web app with everything I want with minor issues to track and fix.
GPT-5.6 didn't try to fix the bugs (as instructed) but it did surface them, which is something that didn't happen with GPT-5.5. When spec/plans approved Fable sends back to GPT-5.6 for ralph implementation and it seems to be an even faster, more reliable workhorse than it already was in GPT-5.5. Overall, impressed. Will continue to be a core piece of my workflow.
While in the past months Composer 2.5 was a lot better than I had expected a year ago, and the GPT 5.6 family does a good job in terms of cost for performance, I wonder why nobody is talking about Grok 4.5 high? Those numbers look very convincing to me.
Other than that, I think the difference between 5.5 and 5.6 will be the same as 5.4 and 5.5. 5.5 is just less frustrating to use, although not perfect and still has derp moments. But a lot less than 5.4.
So I expect 5.6 Sol to be smoother to use. But so far it just feels slower. We'll see.
This kind of "out of bounds" image analysis seems to be a very difficult problem to solve, but totally necessary for transformers to really bring about massive change.
I’m explicitly telling it to do something extremely specific and it’s just not listening to me.
Eg, I gave it an image to update. The image is sized 400x200 pixels. It then generates a new image at 300x300. I explicitly state to be 400x200 in size and it won’t listen.
Sounds great.
Also latency looks very good.
I follow a sort of waterfall workflow which is verbose but fully transparent.
Anthropic’s $100 subscription works fine for me, but whatever subscription my company has with OpenAI reaches the 5hr limit ridiculously quickly.
That leads me to a question. Why wouldn’t they just default to terra in ChatGPT in the last few months? If they didn’t then they burnt money for no reason by giving a shittier model at a higher price
Fable or Opus would wander and wander.
Am I way off base? Seems astronomical.
Or ... maybe Gemini 4 is too good and the NSA is using it to break into systems worldwide ...?
> That advantage extends across the family: Terra performs just above Fable 5, while Luna outperforms Opus 4.8; each does so in roughly one-third of the time, with about half as many output tokens, and at approximately one-quarter the estimated cost.
Wow. I don't believe it. Every indication and twitter post told me that Fable is much more intelligent than Sol and here we are told that even Terra outperforms Fable?
Not only that, Sol doesn't even come with run time classifiers. So it is even more suspicious.
What's even stranger is that OpenAI is directly referencing a competitor in this direct way.
I wonder how long model size and effort will be a few discrete points instead of continuous.
Its an extremely capable model. I think the way we need to approach works shifts again. We need to get our harnesses/workflows to let it gather some momentum on the first couple rounds but then we also need to structure it so that it can slingshot and accomplish the long range goal.
Switching next month. Looking forward to working with Sol.
> GPT‑5.6 is priced per 1M tokens across three model sizes: Sol is $5 input / $30 output; Terra is $2.50 input / $15 output; and Luna is $1 input / $6 output.
Just as expensive as Fable 5. But of course, another slot machine upgrade but the costs will keep going up and the open weight models from china will continue to race everyone else to $0.
Looking forward to the next version of GLM, Qwen, Deepseek and Minimax.
Really wanna see it in DeepSWE benchmark
At work, it was able to one shot a dashboard. Of course, my prompts are vague as I'm not exactly sure what I want yet, but it did a better job than I could do as a backend dev forced to work on frontend sometimes.
Usage is also great, it just feels so much more efficient than older models in terms of thinking and time. Cost is barely better though.
It can burn a million tokens in less than a minute, at least at launch where there's likely less load on the servers.
At home, it feels like I'm fighting the AI less while letting it refactor code. I'm glad that I left this 12,000 line vibe coded port of a hand written codebase to future models to refactor. It feels like the model has better judgement than old models that would destroy your codebase so long as it meant accomplishing your prompt.
I'm almost disappointed that it's this good.
> Reload to try again, or go back.
This on iOS, safari
Great to read they are moving away from the 5 minute cache defaults. Hopefully other providers follow soon!
The good news you don't have to send your dollars to China to fund ai dictatorship, in russia, north korea, african countries and south america.
Grok 4.5 is interesting because it's smart enough at great price. It seems gpt 5.6 is right there with great efficiency and great pricing.
Working with Fable has been a great experience, but at the end of the day, if you can get only 10% of your work done because it just burns through tokens, that's not that interesting.
I've been mostly using Opus and Fable high for planning and codex 5.5 medium for implementations. Claude is also the only model i can use for design tasks. If gpt 5.6 can finally deliver on the design side, it might be time to ditch the Claude sub and go full Gpt.
for one thing, they said that on AA, sol is "within one point of fable" at 58.9 vs 59.9 but don't clarify that the latter is with safeguards where ~8% of the tasks got routed to opus
i'm not rooting for either and genuinely think that the token efficiency and cheaper price are important but this sort of thing just feels disingenuous :-/
At least give me the article ffs.
SWE-Bench Pro Sol: 64.6% Fable: 80% Opus: 69.2% (!!!!)
So, it still trails Opus, significantly, and is not a next-gen coding model like Mythos/Fable 5.
Disappointing to say the least, but somewhat expected.
Open weight models being 10x or more cheaper is just so much more of an unlock than incremental gains for me.
Never go over the free limits in Gemini Pro.
Gemini is great at research and architecture, and my 30 years experience in programming everything; for fun or work; means together there is little to no code slop.
Add to project repo some git submodules of reference source code; boom, bobs your uncle
Zero reason to sign up for OAI or Claude. With employers realizing the costs are more than employees, local models getting more powerful, and models in chips just a few years out, neither of the one note LLM companies without diversified services and R&D portfolios gonna last
15 hits
Holy shit. They must be feeling very threatened by Fable if they're spending this much energy talking about it in the release notes for their own model.
ChatGPT 6 must be deep in the pipeline and will be released within the next few months. Maybe that's why this release is versioned 5.6, not 6.0.