The benchmark comparison to other models was suspiciously missing. The self-comparison is a good representation of progress, but it is light years behind frontier models. Speed is great, but wrong is much worse than slow, IMO.
Yeah of 10 minutes ago. It is shocking how long some seemingly simple things can take. I know there are some things I can do faster than the LLM and some things it can do faster than me. The amount of rambling BS is the exhausting part.
Try different models, it's a breath of fresh air. GLM 5.2, etc. all make life much more enjoyable. They may not one-shot a complex project the same way that Claude can spit out memorized architectures, but that sort of system is always only useful for a one-off prototype anyway, so not much is lost.
I think lithium air batteries will be the real inflection point for aeronautics. Li-air has potential density 12 kWh/kg. CATL is focused on this tech. May be some years before they are able to achieve the full potential density. Still, as the other commenters said, the efficiency of electricity make make up that difference.
yes experimenting with it actually, will update here! in fact we've generated most of the tourist spots in sf, should be reflected in the repo soon too
I really wish GLM models had vision capabilities. I've worked around that in the past to use a vision MCP in my harness that GLM can call. It is not the same, but it allows the model to query images.
That's wonderful. I was going off an older version of the Artificial Analysis page for GLM-5.3-Flash https://artificialanalysis.ai/models/glm-5-3-flash. The page is updated now to show that it does support multi-modal image inputs.
Give it a couple days, and there will be plenty of other inference companies hosting it. Don't like z.ai's TOS? Use the model on a provider with TOS that you agree with.
The pain points in the article do not bother me. I'm bothered by Opus 5's verbosity. It is so long-winded and you have to read through verbose outputs to mentally distill what is important. It is exhausting. I don't think I've ever started skim reading LLM output more that I do with Opus 5. I use the caveman skill, and I think that does help some.
There is so much opportunity for purpose built models like this. Ideally a harness should spin up a subagent to offload to targeted models for specific tasks like this. I know this is not a novel idea. Claude code does some of this by handing off the "explore" agent work to haiku. I just love seeing that specialized LLMs are being developed.
> Claude code does some of this by handing off the "explore" agent work to haiku.
That is not handing off to a specialized model, its just handing off to a lighter and interior model (compared to the parent model). That by itself can create issues like the lighter model not capturing all the data that the parent needs.
The idea is that we get specialized models that are better then general purpose models. But its rare for a specialized model to beat a strong general model.
There is a reason why we hear less about this idea of smaller expert models, because large strong models to the tasks just as good.
And if the tasks is repetitive to the point that specialization is useful, you can get into a situation that your better off having a program written for that reputative nature, then delegating to other models. And then have the main strong model, deal with the (semi)cleaned up data.
> There is a reason why we hear less about this idea of smaller expert models, because large strong models to the tasks just as good.
Smaller models are cheaper, sometimes faster. I agree that the “we’re an LLM fine-tuned for X” hasn’t worked out because you can just train Claude to do X (and Anthropic will), but not burning Opus/Fable tokens on dumb-but-token-heavy tasks is good sense.
As we move from “integrate AI into Y” to “optimize the ROI on Y”, we’ll see more of this.
castform founder here. the roi optimization makes sense. i think there are lots of usecases for which even a 2% gain in accuracy can be quite useful. off the top of my head
- high volume customer support. higher accuracy means fewer escalation, reducing labor costs
- fraud detection. catching even one extra fraud attempt could mean a lot in savings
- and ofc the classic ads use-case where at scale bps in improvement could mean millions in revenue :)
the bigger model would still cost more :)
at the same time, i see prompting as being orthogonal to post-training. i'd imagine post-training a smaller model with a better prompt would make it perform even better
> The idea is that we get specialized models that are better then general purpose models. But its rare for a specialized model to beat a strong general model.
Are you sure about that? I mean, MoE is basically an array of specialized models. This beats single-model general purpose performance.
Clever routing systems also seem to indicate frontier or frontier-beating capabilities at reduced cost, such as Fugu.
Further to this though, I think it hasn't been pushed hard enough. We're not going down far enough in the abstraction chain. Most peoples idea of 'specialized models' is an LLM trained on a particular domain (like math vs coding vs creative writing), where-as it should be going lower-level than that and incorporating not only more abstract and smaller specializations of LLMs, but entirely different types of AI models, classifiers, etc.
I also think the reason we don't really hear much about this (yet) is because there's no incentive for the best AI researchers/labs to go down this direction because it's better suited for local-processing over your entire codebase, but the labs all want you to perform inference on their servers and spend your money there, rather than spend your own compute to increase intelligence. They have a larger incentive to try and scale in different ways. We'll have to rely on open-source for this one.
You can register models with mcp. I think it’s an expensive solution, but it is available in the framework. I use a light weight bus protocol that lets agents interact and pass short messages with pointers. It’s very efficient.
castform founder here. while it's "rare for a specialized model to beat a strong general model" today, i think the tech/knowhow on how to do so is getting there. we see some early signs of this in industry e.g.
at the end of the day, models are only as good as the data they're trained on. and if one has access to proprietary data, it should yield specialized models that do better than larger general models
There are! Chroma has Context1, SID has SID-1, and you'd actually be surprised at how easy it is to post-train your own with pretty good pass@ recall@ ndcg@ etc.
There's also Hornet who have shared some interesting talks & blogs lately. I don't know that I'd exclusively use agents for retrieval the way Neon outlines here as well. I think distillation similar to what ZeroEntropy has done for bespoke retrieval & reranking with _some_ agent manipulation on top-k results works better (IME).
By post-train I presume you mean a finetune? Unless that's wrong (please correct me if so).
I haven't looked into model architecture people are working with for this stuff too deeply yet but I presume the core idea is fine-tuning a lightweight reasoning-enabled LLM specifically using search as a metric for training?
> Claude code does some of this by handing off the "explore" agent work to haiku
This is no longer necessarily true. As of 2.1.198 [0] (released July 1st): "The built-in Explore agent now inherits the main session’s model (capped at opus) instead of running on haiku"
I feel like the future is people building applications with tightly integrated LLMs that work hand in hand with the application's own lifecycle and code.
I also didn't realize that people were using agentic harnesses for search, it's an interesting idea. If the context length is short enough it should be fairly cheap compared to running "normal" agentic coding workloads where you have O(100k) context length for doing almost anything.
castform founder here. that's a future we are really excited about too :) ideally, you can post-train the llm within the application itself, as it's being used. both interesting infrastructure & algorithmic challenges here
Yep. It's a real shame that the labs are incentivized not to go in this direction. They all want to try and suck us into the cloud and take away full control and local processing, but there's far more opportunity by building small AI systems and tools into the harness itself to make the models more intelligent.
They naturally don't like this direction, because it draws the intelligence away from their systems, and onto the local machine, where idea moats cannot be protected and hidden, and costs can be dramatically cut. Imagine though, how powerful our harnesses could be if the best researchers were thinking about how to utilize the power of the gaming GPUs that most PC users have (or can get), to supplement the frontier model processing. Instead of trying to have the frontier model do everything, the frontier model can serve as the orchestrator over all of the smaller dedicated harness models. Right now my rtx4090 sits there unused for most of the day while I'm paying for inference in the cloud... It's such a waste of parallel intelligence bandwidth.
I'm not just talking about LLMs either, most people seem unaware that there are a plethora of dedicated AI models for all sorts of conceivable pipeline usecases, from all sorts of classification tasks all the way down to things like code duplication detection. Right now the LLMs completely suck at cleaning up code and architecture, and a big part of that is because the frontier LLM cannot fit your entire codebase + all of its long-chain reasoning into the context window. But using small local models and tools bypasses this problem because small fast models can iterate over an entire codebase quickly. A harness that creates a big model bundle + routing system + DAG-based memory/execution management over all of these has the potential to be incredibly powerful.
Even better, building a framework around this concept and having the frontier model dynamically and adaptively generate the ideal execution system for any given task/domain. We're working on coding today? Okay, here's a recipe we can use: ..., and it generates a local model pipeline execution system that it feeds all of your prompts through in real time by using pre-defined or shared recipe building blocks, etc... Lots of interesting possibilities.
There has been an over-obsession with frontier models and benchmarks. Most of the work will be done by task specific models. You don't put Phds on the factory floor.
If the PhDs don't cost very much more than your equivalent of factory-technicians but still get the job done, why wouldn't you do that, at least in the blunt case before cost control rears up?
IMO, purpose built or “adapted” models are The Next Big Thing. If I’m using a model to write Python code, for instance, I really want the 8B or 27B expert model for exactly that, which would also be runnable locally. I don’t care about the 1.8T model that can answer every query under the sun and that only runs in a remote data center.
It is nice to have a model that can "do it all", though. And surely that's still the end goal? Like how MoE is still somewhat popular in certain areas even after its heyday.
I am wondering if models will end up being some sort of evolution of MoE where it has something internally like the model the author refers to that gets surfaced when it needs to search in some way. I guess it makes sense; our own brains have so many distinct task-specific regions.
Yes, agreed, but I think it’s going to be difficult to have a high quality model that “does it all” and have it be local. For a quality “does it all” model, you’re going to need a lot of parameters and that means you’re almost always going to be running in the cloud. But it seems like you could probably get a smaller, focused model that runs locally and is also high quality. In other words, I want the programming expert portion of a 2T model that is maybe 35B or 70B parameters or whatever but it’s running locally (and yes, I know you don’t just carve out an expert from a larger model, but conceptually that’s what I’m after).
Yeah, I always thought the future of this stuff would be hot-pluggable MoE modules or LORAs that are able to be downloaded and applied/used at will like how skills have become a thing.
Like, atm most architectures seem limited by a single context and fixed architecture with no hot loading. Especially for robotics, being able to load/unload various specialised skills on limited mobile hardware will (I hope) definitely become a thing.
(one of the blog post authors here) -> once you set up a finetuning pipeline, it's often trivial to rerun it on top of a new open weights model. so, it's orthogonal to base model improvements
> There is so much opportunity for purpose built models like this.
OpenAI etc could themselves do this, and maybe they already do? Where the public-facing interface delegates to multiple little goblins behinds the scenes
Exactly. There could be a lot of value for inference companies to do this. Could save a lot of money being able to hand off highly repetitive known tasks to far smaller specialized models.
isn't this also the threat to frontier AI houses? As in they want you to expend tokens in their ecosystem, but the optimization at 100x is their profit?
castform founder here. i'd say it's a threat but the frontier ai labs' argument would basically be that the market opportunity for intelligence is infinite so it doesn't matter. at the same time, i do believe there will continue to be a massive, growing market for big labs, esp for super-frontier use-cases. today that's longer-horizon coding tasks but in the future it can be things like scientific discovery, etc.
castform founder here: totally! we also think model routing is also a post-training problem i.e. getting a model to predict the difficulty of a task and match it to the right model -> we're gonna be sharing more on that soon :)
I’ve been trying to do this with a pet project and admit that I’m getting terrible results. My small llm as a classifier/router stage just isn’t getting the job done.
This looks fantastic for a common async workflow I use. I often use one job to fan out multiple individual http request jobs. The reason I prefer jobs for this is easy and consistent retry logic, and durability. I want to make sure those HTTP requests eventually go through. Fibers would be much better suited for this. So much of work that goes onto work queues is IO bound, and fibers are a great fit for that.
reply