GPT-6.1 Terra
openai.com543 points by meetpateltech 10 hours ago
543 points by meetpateltech 10 hours ago
https://developers.openai.com/api/docs/models/gpt-6.1-terra
Similar 75% price cut trajectory as GPT-5.6 Sol -> GPT-6.1 Sol. Overall, Terra is still half the price of Sol. But like the original release of Terra, it's still a weird middle child between the small-model-that-got-substantially-better (Luna) and the large-model-that-got-significiantly-cheaper (Sol). The TerminalBench 4.0 cost-per-task results are a bit confusing and I'm not sure why you would pick Terra over Sol in that case. Overall this release seems like a flex on Sonnet 5.5: a model which is having its own middle child problems between the godly Opus 5.5 and the surprisingly capable Haiku 5.5. Artificial Analysis still has MiMo 2.6 Pro setting the Pareto frontier where it matters: it's as cheap as Terra High/Sol Low while being smarter than Terra Xhigh. Sol Medium beats MiMo in smarts but is priced a lot higher. Meanwhile Terra Max is dominated by Sol High in both price-per-task and smarts. the surprisingly capable Haiku 5.5 Destroyed by Anthropics crazy decision to price it x5 after 100k context. do people really use luna post 100k? small models generally seems to be hallucinating hard for me once you are over that threshold. I don't think software devs are the audience for 6.1 Terra, but instead computer work, automation, and technical writing. Honestly I'd much rather take this model then. It is having improvements in other fields than coding while being more cost efficient. If all I want is coding at a cheaper price, then deepseek is a good model for that. I am more than okay to have a bifurcation of models good for coding and models good for everything other than coding. I strangely imagined it many years ago. Everyone is suggesting OpenAI to release a model close to opus 5.5 yet it's releasing this. On one hand we can imagine that maybe their training cycle wasn't able to produce this whereas gpt 6.1 terra was made. But on the other hand, I also now wish to see an OpenAI model which is good for just pure coding, like an absolute beast at it while being cheap (so the opposite of terra) ironically, one of the few appreciations that I might give to OpenAI if they pushed this model to prod and are facing some pushback as developers like us want/only see coding models. They could've benchmaxed it and it would've been slightly irrelevant. I'd much rather see reality. At the very least, this new model won't make things more expensive for anybody and would only make use-cases cheaper depending on what they are, so it's a fine deal for me. I also wonder maybe this is why open source labs aren't targeting this space and why the only focus of AI labs is coding nowadays. Everything else just doesn't seem so worth it from investor, even end user, training perspective. Maybe this is the reason why we have gotten almost an exponential growth in coding capabilities as that became the most (if not only) capability that truly really matters. To not wish for discussion to go to a tangent, nonetheless this could also explain why our work as a software engineer is going through a slight crisis of meaning as these AI models have gotten impressive at coding. As an (aspiring) developer (currently in 1st year college), I don't wish for AI to replace coding even more for me or to push it even more or to suck the joy out of it even further. So I am more than happy for these models to stay as they are and rather make advancements elseplace. Ironic because it is us developers who use AI as well and we want cheaper workflows and so we would also want models to be cheaper and better at coding. There are some fine lines of contradictions and irony in all of it. I'd much rather also have open-weight models of such terra-like models but I imagine that because of the reasons that I mentioned above, we are most likely gonna see it the most focused on coding but still open-weights is pretty close. The more choices the better (though choice-fatigue is a real thing), I hope I was able to explain some of the nuance I feel about this particular launch. I'm very confused about their model market position/naming schema. First we had: > Luna (Small), Terra (Medium), Sol (Big). Which was fairly smooth after you got used to their planets. But then they added Astra as the "Bigger," which was mostly still self-coherent: > Luna (Small), Terra (Medium), Sol (Big), Astra (Bigger) But then things go confusing, they released Sol-6.1 which sometimes is superior to Astra, and discontinued Terra which made it look like this: > Luna (Small), Sol-6.1 == Astra But with Sol-6.1 costing the old Terra-6.0 price, while, as I said performing as good as the old "Bigger" model in many non-3D/UI situations. Now they're re-introduced Terra, as a 6.1 release, making this even more confusing. Did Microsoft's chief branding officer move over to OpenAI? I work at OpenAI and I can try to explain our logic. In general, LLM performance is incredibly multidimensional. There's no guarantee of a fixed ordering where the model that is best at math is also the model that's best at jokes. Despite that complexity, much of the variation in performance is explained by just a single variable: model size. In general, models that are bigger will (a) be more performant / reliable, (b) cost more per token, and (c) have better recall of esoteric facts. As you note, our models are ordered: Luna < Terra < Sol < Astra Even today, I'd recommend this as a general starting point for picking models. If you care about quality only, pick Astra. If you're a bit price sensitive, pick Sol. If you're very price sensitive, pick Terra or Luna. Of course, model size doesn't capture all variation. Models that are trained later might have new data or new techniques or new architectures that earlier models lack. You can see this with models from all providers - Sonnet 5.5 outperforms Fable 5.1 in ways, and Gemini 3.8 Flash outperforms 3.1 Pro in ways. In the case of GPT-6.1 Terra, I'd say Terra is probably better than Astra and Sol at technical communication, similar on vision and computer use (but much faster!), and a bit worse on most other things. For API developers, it's a nice midrange option if you want something cheaper than Sol but a lot smarter than Luna (especially for vision / computer use / communication). For subscribers, it's not a bombshell release - if you're hitting usage limits, maybe it helps you go a bit further, and if you want clearer communication or faster computer use, it can be a nice option to select. If you have suggestions on how to name our products better, I'm all ears. Our intention is to make things clear and simple, even if we miss the mark sometimes. Unfortunately, a lot of the complexity is somewhat irreducible and due to to the actual complexity of the products. Scaling model size is different from scaling reasoning effort which is different from scaling multi-agent count. So if we just swept all of those under a single name of "more IQ", it would make the naming simpler but the experience much worse, as important details would be obfuscated from end users. Do you then recommend Sol for technical documentation? What's confusing to me in addition to the naming is that there are no published guidelines I can use, as far as I can see. Put yourself in the user's shoes. On what basis should one pick a model for a given task? I can understand smaller models being dumber and faster. That aligns well with scenarios where you are willing to compromise; e.g. you're classifying content or asking a model to format text into a table. But when you want "maximum intelligence, cost and time be damned", there's the implication that bigger models are smarter on a linear scale, and every provider's model release naming is currently conflicting with that expectation. This isn't just a naming problem, but a communication problem. If I knew that Astra was the best at coding but a bit bad at writing, and Sol was almost as smart but better at technical writing, then that's the kind of thing I expect to be documented. One possible way: A simple matrix with horizontal columns like "writing", "coding", "math", rows are the models, and each box is a number of stars from 1-5. Instead, this sort of thing becomes lore, shared by word of mouth. > Do you then recommend Sol for technical documentation? I don't know, to be honest. Technical documentation is such a multidimensional skill, and I also don't know your budget. I'd imagine Terra might have the most readable prose, but Astra might be more likely to catch bugs or mistakes to highlight. If the documentation is about a more esoteric language or situation, Astra might have a greater edge than if it's a common language or situation. If you have a long list of documentation guidelines, that might also give more of an edge to Astra. I wouldn't want to make a recommendation without doing a lot of testing, and I haven't done a lot of testing. Some food for thought, if someone from oAI can't make a direct recommendation for something as direct as "Technical documentation is such a multidimensional skill" - how could we know assess what model to use? To elaborate, I can ask each model to write a doc about something I know well and tell you which output I like better. But to assess whether my preference on my problem is going to generalize across other people, other programming languages, other documentation types/formats, etc., and then analyze statistical significance is a thornier question that I don't have the data to answer. I'd rather make no recommendation than a weakly informed recommendation. I use Astra 6 Extra High for everything. It’s overkill most of the time, but frankly all these models are “cheap”, and I’d rather waste compute than brain cycles trying to discern which dials are best for each individual as-hoc query. How are they cheap? Does your organization offer unlimited spend, or does your subscription not run into the same limits as mine? Maybe if you have subsidized 200/500 dollars subscription it may look that way. > Put yourself in the user's shoes. On what basis should one pick a model for a given task? Create a realistic eval suite for your use case using your actual data that you manually label, with easier, more complex, rarer samples, and run each model you are considering against it and decide based on the usual confusion matrix, cost, latency, or whatever metrics matter to you. There's really no other way to know – and you might be surprised that cheaper models can often perform your task(s) relatively well at a fraction of the cost. > Create a realistic eval suite for your use case using your actual data that you manually label, with easier, more complex, rarer samples, and run each model you are considering against it and decide based on the usual confusion matrix, cost, latency, or whatever metrics matter to you. I’d imagine there’s only a tiny percentage of users that can or want to do this. > Do you then recommend Sol for technical documentation? Every model I've tried is just abysmal at this, for what it's worth. If any do it well please let me know and I'll give it a spin. My thoughts exactly. I recently experimented with some image generation models and I was surprised at how much context/prompting is needed to get something that doesn't look Lovecraftian. There was seemingly no downside to adding more and more detail up to the limit. Prose sits between code and images in terms of fuzziness and I wonder if promptmaxxing is the winning strategy for technical documentation. I've been thinking that fine-tuning a model to write differently shouldn't be that hard (tedious, perhaps) and if I had the resources I would consider doing it myself. You can start with Artificial Analysis benchmarks but, then you need to test in the real tasks anyway. I assume having a person write words for other people to read is not feasible? Humans cannot be in the loop for autonomous work. I have agents review code and design/architecture documents, write alert messages based on rules, write analyses (e.g. outage incidents), and many other things. "Having a person write words" would mean not using agents, defeating the whole point. I do not use agents to write on behalf of myself. I write my won technical write-ups and my own slide presentations. A lot of people don't; and it shows. There's much that agents can do better than humans now, but not everything. This isn't technical documentation though, it's just notes the agents are keeping for each other. I think that is a very different task > I'm all ears. Our intention to make things clear and simple, even if we miss the mark sometimes. Thanks. I think the very response you wrote here is a better form of marketing for the software engineers than what openAI puts out there. When everything is groundbreaking and everything tops charts, nothing is. I suspect a lot of target demographic (developers) prefer simpler and honest communication about not just the models high points but low points too. With the linked Terra post, if I saw 1 chart that clearly highlighted the usecases for which Terra is _worse_ than higher tier models, I would be more convinced overall. Everything trying to occupy the top spot in my brain's limited "top model" representation space is not helpful. While I generally agree that it'd be nice to just have "here is the Pareto optimal model for your use case. Here is the specifics we optimized each model for" so users don't have to do benchmarking and research as much, they're getting much better at explaining. It's basically what the commenter above said. The website linked starts out with: > We’re introducing GPT-6.1 Terra, a fast, efficient model that combines detailed visual reasoning and precise computer use alongside clearer, more natural writing. A significant step up from GPT‑5.6 Terra, it excels at the communication and computer-based tasks that fill the workday, from drafting proposals and interpreting technical drawings to filling out forms and editing videos. If you go father down, it has graphs and details with specifics of this, including: > We measured GPT-6.1 Terra’s computer-use capabilities... At medium reasoning effort, GPT-6.1 Terra scores 15 percentage points higher than GPT-5.6 Terra at about one-third the cost. It even performs on par with GPT-6 Astra at only one-tenth its price per token. and > On MMMU-Pro... It even nearly matches the scores of our most capable frontier model, GPT-6 Astra, and GPT-6.1 Sol at much lower cost. The 4-tier naming is honestly as clear as you can get. I have no idea what people are so confused about. Terra and Luna are dirt cheap and good enough for 50% (or more) of the typical junior dev's workload, and great for routine work. Sol and Astra have so much written about them at this point that it should be obvious what those are good at. The large/gigantic models for things that need full intelligence and slightly more branching thinking. The obnoxious/confusing thing is that models are released individually rather than as a family. Luna < Terra < Sol < Astra, fine. But does Terra 6.1 beat Sol 6? Does Fable 5.1 beat Opus 5.5? If I build something around this Terra 6.1 release, then when Luna 6.2 is released in a month, should I swap? > Luna < Terra < Sol < Astra, fine. But does Terra 6.1 beat Sol 6? Does Fable 5.1 beat Opus 5.5? If I build something around this Terra 6.1 release, then when Luna 6.2 is released in a month, should I swap? I understand how it can be confusing, but I don't think that this is exactly a problem. Luna through Astra offer qualitative descriptions of the model's size. Presumably OpenAI trains its biggest model (Astra-ish, maybe even larger) first, then distills it to smaller sizes. Verison numbers reflect something about the architecture or training dataset. Again, they keep these details close to their chest, but we should expect that weight-for-weight higher versions should beat lower versions. This offers us two axes of 'better', but these axes are also not directly comparable. An earlier but bigger model may or may not beat a newer but smaller model. This is where 'spiky' intelligence comes in. This is also where calls to 'run your own evaluation' really do start to matter. If you care enough about this performance[1], then you also have specific problems that you care about. No generic benchmark can fully answer that question, hence the need for a specific evaluation. [1] — And you don't have to! A casual user is probably served reasonably well by anything put out in the past six months. I think Microsoft is still serving GPT-4o under some Copilot branding. I think it is a problem, for OpenAI. If their answer to "which of your models are better" is to throw their hands in the air and say "fuck if we know, go figure it out yourself" then that's going to go poorly as soon as they try to actually sell to some of the market they're predicting they can sell to. Whether they train the biggest model first or not is not relevant; what matters is what they release. Right now we've got 6.1 Sol and 6.0 Astra. Over at Anthropic they've got Opus 5.5 and Fable 5.1. The existence of a hypothetical Fable 5.5 that I'm not allowed to use doesn't help me. I understand the confusion, but isn't this kind of similar to asking does the iPhone 14 Max beat the iPhone 16 Mini? At least with phones there are hard specs we can use to compare, but in a lot of cases the answer will be "it depends" At least there is a simple system - the words 'Mini' and 'Max' are intended to refer to the size of the item (that one is miniature, and the other is their maximum size). If they were called iPhone 14 Luna and iPhone 16 Sol then this is less clear. Or on the flipside, if the models were called GPT 6.1 Mini and GPT 6.1 Max, nobody would be arguing about the names being hard to grok. The words 'mini' or 'small' are more descriptive than 'Luna' or 'Terra'. I guess it never really occurred to me that people wouldn't immediately grok: * Luna (Moon) * Terra (Earth) * Sol (Sun) * Astra (Galaxy/Universe) The relative sizing seems dead simple. You're misunderstanding the misunderstanding. You're right! The relative sizing of models is dead simple. This doesn't help when there are improvements between release numbers, and models are released one at a time instead of as a family. When Sol 6.4 comes out and the latest Astra is still 6.1, which do you use? I use Sol 6.4 because it's newer (and "big") unless there is a specific benchmark that 6.1 Astra scores better on that I think I need. Or I might try both and pick whichever I prefer for a given task...
minimaxir - 10 hours ago
zozbot234 - 3 hours ago
rockdoe - 6 hours ago
michaellee8 - 2 hours ago
wyre - 9 hours ago
Imustaskforhelp - 8 hours ago
Someone1234 - 10 hours ago
tedsanders - 9 hours ago
atombender - 9 hours ago
tedsanders - 8 hours ago
flockonus - 5 hours ago
tedsanders - 4 hours ago
scarab92 - 4 hours ago
selestify - 4 hours ago
hsuduebc2 - 3 hours ago
lazharichir - 7 hours ago
awb - 5 hours ago
CSMastermind - 7 hours ago
zenoprax - 6 hours ago
ilaksh - 8 hours ago
saagarjha - 7 hours ago
atombender - 7 hours ago
saagarjha - 6 hours ago
GodelNumbering - 9 hours ago
vineyardmike - 8 hours ago
elxr - 9 hours ago
margalabargala - 9 hours ago
Majromax - an hour ago
margalabargala - an hour ago
cdirkx - 9 hours ago
Closi - 8 hours ago
rpeden - 3 hours ago
margalabargala - 2 hours ago
arcfour - 28 minutes ago