Qwen3.8-Max: A New Bar for Coding and Cowork
qwen.ai
qwen.ai
Particularly, I had one team member who was extremely sceptical of AIs/LLMs/harnesses and refused to use them. One day he said "Well, I have an RTX 5090 doing nothing... should I try to get something up on it?" and a few minutes later he had 3.6-35B loaded up, running OpenCode.
It continues to be a workhorse to this day, running on both my local Mac for various types of jobs, an AMD R9700 at the office, and said teammember still uses it on his 5090, although in practical terms we do a lot more with DS-V4-Flash-0731 these days.
I honestly think that with my electricity prices running qwen 36B myself is more expensive than hitting the cache rate at deepseek.
Kimi K3, GLM 5.2 and now Qwen3.8-Max - open weight models.
DeepSeek V4 Flash outperforming Gemini 3.1 pro, probably DeepSeek V4 Pro update is also coming soon
Chinese labs are cooking very hard. US closed weight labs are probably hard time to resist not calling Washington DC for more AI regulations
Not sure how Qwen3.8-Max is going to be licensed, hopefully it'll be Apache like the smaller ones.
There seems to be quite a gap between the small ones and the enormous ones these days.
Just so that we know what 3.8 would be like.
I currently have about 150 Tabs of Antirez posting on AI and running local model I haven't had the time to read. And there are probably some prerequisite reading or other research in between as well. I just wish there are some very high level overview and news coverage on all these.
IMHO this is a difficult question to answer. Part of the power of paid models comes from the software supporting it. With local models, you have tons of workflows that can severely influence the quality of the result.
In my personal experience, the SOTA models are way more consistent and can handle more complex questions. Part of that is (probably) because I don't let my local model access the internet, while paid models do use the internet to look at docs etc.
I guess which of the smaller 3.8 models is best for coding will depend on which one they put the training effort into.
https://humanparadox.org/local-vs-frontier-benchmarks-for-my...
It can complete multi-step tasks much better, and has a bit more curiosity.
I believe I am competing directly with these frontier models in some circumstances. Like there are a ton of programmers who previously would be outsourcing work to that site, but now they assign that same work to AI agents.
Ever since November 2022 when ChatGPT blew up, I have been focusing on agents in order to try to get ahead of the curve. But I haven't managed to get an agent business off the ground and have been doing poorly paid agentic projects from that site instead.
But now everyone is building agents, and this crazy list of accomplishments makes it look like we are close to the point where the agents are building agents.
In fact the next time I get an Upwork contract for another agent, I actually should run it through my agent and see how far it can get. What I'm seeing a lot of now is requests to automate as much of a business as possible.
Anyway the point is these models are just about capable of doing the entire job of analyzing a small business and building out all the agents and iterating on them with the business owner.
That's actually what I should build is a SaaS that does that. Which I would if I wasn't basically desperate to get another contract this week.
And I know Upwork is bad but I have not had much success with other options on short notice.
If you search for LLM benchmarks focused on real-world tasks, you'll quickly learn this ain't the case. No point in telling business owners about that though, they need to see/learn by themselves.
The cleanup pay is mighty generous as well.
Still no Artificial Analysis benchmark yet. Or benchmark for Laguna S 2.1 or Meituan models or lots of other models.
I see people, and my self making amazing things with AI and fixing old projects and having real world impact at the fraction of the cost it would take me to hire people, or hours spent on my own coding.
I have built tools and systems with AI that have allowed me to build windows drivers, android apps, web apps, iOS apps, vm occultation, custom block drivers, custom file systems and more. To the point where entire products have been created.
Not trying to be a doomsday, but yes. It seems as though with the right infrastructure we are at the point where businesses owners can go from idea to product very fast and not need or hire much external talent.
You need to run away from any client who thinks your primary purpose is to write code. You must run away from the business if YOU think your primary purpose is to write code.
You should be able to write code and do it well, But the AIs, and I'm not even talking about the ones on the frontier, have been able to write code faster than I have for quite some time.
What you need to explain to your customers is all the things around writing code: software architecture, performance, and so on. Also make sure you have some understanding of the customers business, so you can suggest additional ways to make their lives better/make more money.
Software developers have a chance. Code monkeys? None.
First, I am aware they have been able to write code for some time. I made my first LLM coding agent experiment the day after ChatGPT first blew up in November 2022.
I haven't been a "code monkey" in the last say 15 or 20 years of my programming. I would argue also that code money was never really a thing, it was just an excuse that pretentious overpaid developers used for the existence of underpaid exploited programmers. 95% of programmers have to do requirements analysis and design regardless of their pay grade. I learned a long time ago how important requirements analysis and good design is. And the jobs I am applying to largely specify architecture and requirements analysis in the description. Employers are aware of code generation tools.
But also, the premise that writing the code is not the hard part is just nonsense. Sure requirements analysis is harder than people realize, good design is hard and so is understanding things about proper and closed iteration. But that stuff is not, in the aggregate, harder than the code.
And also, frontier models can absolutely do requirements analysis and architecture. And the sheer speed of implementation means that there is a huge built-in advantage for iterating more effectively.
guess one which is valuable ?
Assume the business logic and all edge cases work in the legacy app. The app is "done", just too old to work on or run on modern hosts. Hence the project.
Would/could you use AI to deliver the project at 10x the speed? Or at 1/10 the price? Or charge the same amount as ever?
I have a feeling the lowest-cost developers on freelance marketplaces are not very good at using the AI tools (and whatever related acumen) and need a lot of hand-holding, the same amount of effort almost as if I decided to guide AI to do it myself.
IMO, there's value you can sell in being someone that (even developers) can hand off a project to, to do end to end, if it's fully specified. Even though the buyer could do it themselves, the developer would be taking on managing it and having AI do it until it's correct, and all the little things, few as they may be.
AI should be giving contract developers superpowers. If it's not, then the limited set of contractors with the AI superpowers will be able to charge 10x as much for their time, and the actual cost of the project remains the same even if it's done 10x as fast/easily.
Curious if you see/feel this, or if there's a marketplace for "please vibe code this for me, make sure it works and is what I want".
I would say 90% of outsourced developers are going to use AI just like someone making top US dollar. Actually for the majority of projects on Upwork, the need to generate with AI is much greater. And I think it's nonsense the idea that an overseas programmer can't figure out how to make adequate requests to a frontier model.
The models are smart enough that someone could literally just say "convert to Rails 8 like you are real real good best Railsdev" and follow up with a few "now test and make the bestest code Base" and if they are repeating that maybe with a few different leading models.. will a real Rails dev be able to find some fault? Sure. Would they have also found issues with the work of another human Rails dev? Yes probably.I just don't think it takes as much actual skill at this point. And certainly there is no secret ability that people in San Francisco have but not in Bangladesh or only certain neighborhoods or whatever.
One guy I was talking to said he was frustrated because his developer was not using AI enough. I came to the conclusion based on his requirements and relatively generous project timeline that that developer was deliberately avoiding using AI or admitting that he was using AI because he knew if he did then the main project phase would be finished in less than a month.
It does help dealing with what previously were often impossible timelines. Recently my Upwork profile has focused on AI agents so I have not been seeing many of those typical line of business type jobs.
But also more than half of those more typical business web apps are now starting with "I have already generated a prototype" and they want it to be made. "production ready" which I think contractors are taking advantage of how vague that is especially since the majority of these projects have pretty modest actual usage. I actually don't like that because it's much less concrete than delivering actual features.
I don't know why but this made me laugh out loud. I know you're trying to help but just a funny jump
I think the volume for custom agents is probably higher than ever but so is the competition for that work.
PS. Not everyone (in fact the rare few) write as if no one is reading. For most, literally no one ever reads :D
My MindRoot framework had some of qm's main features awhile ago though. I do try to use it to build solutions by customizing rather than from scratch when possible. But a lot of clients or potential clients don't like that idea even though it's MIT. They want some that has clearly been invented by them as a new programming project.
I actually turned down a project last year from someone who wanted me to start over in LangGraph -- he had already decided on it before he knew about my thing.
The most recent project I have been using MindRoot and building up the voice capabilities so I can now handle voice agents end to end including SIP.
But on that website now I try not to emphasize my framework at all for most proposals. You don't have leverage on Upwork. And also with code generation it isn't a critical point necessarily.
but then you totally jumped the shark!
All requests to an LLM are idempotent, for every API call you need to send it the entire conversation history so that it can process it. LLMs do not learn or remember anything, which makes it super easy for users to switch LLMs on the fly. Most popular AI frameworks, make this a one-liner change these days.
And that makes me wonder if the trillion dollar valuations for OpenAI and Claude are even justified. Cause if that is justified, then Kimi, Qwen, Deepseek etc are also valued at a trillion dollars. Or all of them are worth a lot less. One of those statements is true.
Also this makes me wonder if the next iteration of LLMs would be based on fine-tuning, where LLMs actually learn from your past behaviour so that it would grant some amount of stickiness to the product. OpenAI used to offer fine tuning runs for GPT-3.5, but they don't seem to do that anymore.
A more appropriate term is “stateless”. LLM responses are certainly not idempotent, as they are not even deterministic.
https://earendil.com/posts/session-portability/
“Just take the session thread to another provider” might not be feasible anymore soon-ish.
This is harder to do on other architectures that themselves aren't fully deterministic though.
Isn't that more due to an optimization and not how the LLM itself runs?
Like a MoE LLM run on a single input should give the same output each time. But this is inefficient, as any given token is hitting 1 (or maybe 2 or 3) experts at a time, meaning all the other experts are doing absolutely nothing. So you upgrade it to take in multiple requests. But then any given expert can become a bottleneck, so when too many requests need a given expert, some of them are routed to a second or third best expert instead. Within the context of any single request, this looks like non-determinism, but it is still deterministic when considering the full batch.
For everyday users and everyday use cases, that is enough to treat it as non-deterministic (the harness might also send in unique data like current time which means one can never have the exact same request twice), but when talking about LLMs more theoretically, I think we need to consider they can still be ran deterministically even if that isn't as optimized.
Similar with temperature. 0 means deterministic, but anything higher with a seeded value is deterministic. If anything, temperature is us purposefully adding non-determinism to agents because they were too deterministic.
https://www.semianalysis.com/p/google-we-have-no-moat-and-ne...
They aren't, not even if we forget about the capable Chinese models.
I suspect Anthropic will implode soon when employees are unable to get the cash-out that they expected. Having so much compensation locked up in company stock is risky on a good day.
The first is the compute. OpenAI and Anthropic secured huge amounts of compute, Google, Meta and xAI have their own huge datacenters. Now anyone can rent some cloud machines and start serving Kimi K3, but it's going to be impossible to get to a similar scale as the big 5 above. And inference has economies of scale: the more people you serve in parallel, the more efficient you are.
The second is the data. By now (and maybe even by one year ago), all the data on the internet has been used for training. You need new data. The big AI companies sit on top of trillions or quadrillions of tokens that they have generated over the years. They can use that to train new models. That data is gold, and the proof is that SpaceX was happy to pay $60B to acquire Cursor.
If you want to overtake the frontier labs, you have 2 options: use their models to generate synthetic data, and provide lots of (cheap, maybe below cost) inference to generate your own new data. The frontier labs know about the first, and I'm sure they try to limit how much others milk their models. As for the second, that's the "honest" way to compete, but it's not easy.
Will this continue? The field is moving too fast to tell. Kimi, Qwen, Deepseek also produce very capable models but that doesn't automatically translate into trillion dollar valuations.
However, trillion dollar valuations on Anthropic and OpenAI, such new companies, never publicly traded and such huge valuations decided just by investors. This is just asking for trouble.
1. https://docs.github.com/en/copilot/reference/ai-models/suppo...
They don't. The moat will mainly be the tooling around AI, not the AI itself. You don't hear any company claiming their moat is the Internet.
These companies have AI and enough money to lobby the Pope. They can afford to reanimate members of congress and push some tactical legislation through.
But all the money in the world cannot move government too quickly.
Other moats exist too. OS or browser can undermine performance and availability of alternatives.
Generally speaking they do, at least from my experience when switching from one model to the other - their performance decreases, and they often do large refactors outside of the requested scope as they try to bring the code closer to 'their' style.
Which makes sense imo - they'v been trained to iterate over the code they wrote, and not code that was modified by someone else in the interim.
Recursive self-improvement changes the parameters a bit, especially for the market-leaders, and it's the one thing that makes me wonder if they'll be able to extend their lead faster than the smaller labs can keep up, but it's an option available to everyone.
Why is Google a monopoly? Do they have so advanced software that no one can outperform? I doubt it. What they have is a giant user base that generate loads of real-time data, which make Google services more accurate.
So how AI company can build a moat? Exactly the same way: by making a giant user base produce loads of real time data. Just imagine a service that will generate answers not only based on data they were trained on, but on all data from all user conversations. Imagine being at a concert, looking for a certain type of beer and instantly receiving an answer from an AI assistant about that only because some other guy in a crowd looking for exactly the same thing said to his agent "ah, here they are!".
It is not happening just yet because of making it secure and private is not yet solved, but it's just a matter of time I think.
It’s more like a bunch of people are placing different bets. Only a few bets are going to generate a return, possibly only one, but the profit on that one bet will make it all worthwhile. That’s the theory, anyway.
Fine tunes are a possibility but I think it offers very little uplift for the vast majority of uses beyond just stuffing enough context.
Which anyone who hasn't been sitting in the SV echo chamber could have told you years ago after applying even the smallest bit of thought.
Original designs: https://image.non.io/257dc9cb-9e6b-4e00-8f12-23ea5e073649.we...
These are fairly rich, pattern-heavy, nuanced designs. I've asked each to create it as a SPA where the map flows behind it.
Opus 5 results: https://html.non.io/opusAcmeBooks
Qwen 3.8 max results: https://html.non.io/qwenAcmeBooks
Same prompt for both for the conversion. I used OpenCode for the qwen version, but I encountered a significant amount of errors / timeouts while it was running. Claude finished in around 16 min, but I spent close to 2 hours shepherding the Qwen build. For the implementation, there were signs it had good vision, but the timeouts make this very hard to use in a production setting.
That folder didn't exist, I asked Qwen where the files were, and it attempted to create them again.
Stopped it, and asked what happened to the original files. Qwen Desktop apparently doesn't have access to the local file system, but continued merrily along without warning about that.
https://image.non.io/07a153cd-c31c-4f7b-a89d-363faec05a91.we...
It did a self update and it worked for a single request (me just saying hi). Pasted in the prompt to build the page and it 401'd.
I'm quite surprised at the difference.
I've noticed another type of AI slop that's prevalent in all the popular models; font sizes and variations like you wouldn't believe. It's very hard to instruct LLM's not to do this.
There's Mistral. Although they've been surprisingly quiet for the last 6 months.
I agree, but you are underestimating EU regulatory incompetence
Level 2: ... or with anyone who does.
Level 3: Anyone caught with unauthorized model weights is subject to the same prosecution they could expect if they had been caught with CSAM.
However, that would just change the weights values and not their dimensions.
It probably sucks for you if the proposed powerline goes through your house, but the need of the many come first.
Having cheap and plentiful power is a huge market advantage. It was one of the factors driving Norway out of poverty 100 years ago.
The need of the rich, you mean. In these kinds of situations it's never about the many, if it was about them different choices would be made.
EDIT: This was obviously meant to be facetious with the intent to highlight the negative effects on the general population that datacenter buildout has in the US.
volunteers in the open source community use whatever is easiest and cheap.
open weights is not open source. there is no "trust".
I don't understand. That's dated today, but:
https://twitter.com/alibaba_qwen/status/2078759124914098291
> Qwen3.8 is launching and going open-weight soon! [...] You don't have to wait to test it. Just now, the Qwen3.8-Max-Preview made its debut on Alibaba’s Token Plan, Qoder, and QoderWork.
That was on July 19th. I used it to draw this pelican: https://simonwillison.net/2026/Jul/20/afraid-of-chinese-mode...
So what are they releasing today?
I believe they mentioned around the preview announcement that they'd be releasing improvements to capability, which I assume means continued training.
3.8 Max Preview ran 90% discount on QwenCloud exactly as when Moonshot couldn't keep up with all the signups.
Here's a pelican I just got out of the new model. It took 11 minutes and forgot the wheels! https://tools.simonwillison.net/markdown-svg-renderer#url=ht... (scroll to bottom)
The reasoning trace is pretty great:
> More additions: basket with fish in it? Cute detail — a fish poking out of a basket on the handlebars! This adds charm and pelican context.
If the price is $2/$6 that cost me 17 cents: https://www.llm-prices.com/#it=90&ot=29734&ic=2&oc=6
The open models are now good enough for what I want to do with them, let alone any future improvements. And factoring in efficiency gains, a model in the ~70b range starting to satisfy my needs would completely obviate the need to pay others for inference. This does not seem far-fetched to me, comparing with where open models were at this time last year. What am I missing?
So just like in IPhone vs Android, you could end up with a situation where Chinese firms compete and get most of the revenue and usage with low profit margins but OpenAI and Anthropic capture a premium side of the market and still get a lot of profits.
I have numerous cases where Sol failed and only Fable could solve a problem. For example yesterday I was merging a Q2 curved with a Bezier curved face in 3D using OpenSCAD. I tried for over 2 hours with Sol 5.6 high and x-high.
Fable two-shotted it in about 30 minutes.
In my experience open models (or GLM, DS and Kimi) are radically worse than either of Claude or ChatGPT at these tasks.
I think there is a huge "long tail" of tasks like this where the frontier labs are ahead, and I think this long tail is quite important.
Unless you are spending more than a max subscription (200 a month+) its cheaper to use the cloud.
But things are priced cheaper in the cloud now to lock you in and restrictions around hosted models are getting worse.
If you only have a $300 dollar laptop its probably not worth the upgrade.
I'm personally excited by local AI but the experience for the average isn't the same. I'm willing to get .5/s running on 10-15 years old machines but what I can do with it is limited.
Their marketing department :-) . I'm only half-joking; those guys are hard at work finding the best product-market fit for ChatGPT/Claude. "Product market fit" means "strongest revenue", which is not necessarily going to bring the best tool for you or me, but the one that can either get more consumers to shell off money, or more enterprises to cough money for licenses, and in both cases those consumer basis will be narrowed down to what legal and geopolitical circumstances allow OpenAI and Anthropic (and this is why they want to ban the competition!). It also means dark patterns and enshitification, of which I'm already seeing some both in the Codex interface (it was just renamed "ChatGPT"!!!) and in Claude Code (which also is just "Claude" now and can't '@' properly any longer). So in the medium run most people will be better off running an open source harness that can use any model.
you can't create a new drug by running a model on a laptop. You can't serve a customer support bot running on a laptop. You can't generate video in bulk for many users on a laptop. So there is still a case for paying others for inference.
Does it justify the valuations? No idea, but some major use cases are still there. That's why they are rushing to implement, OpenAI creating a "deployment company", Anthropic having some pharma rumors, etc.
LLMs are useful. We can all see that in agentic coding. But replacing everyone’s job? Hardly. And what’s with the scare tactic of trying to get the US government to ban foreign models?
LLMs are useful, and dare I say they’re on par with the internet. Making them cheaper and affordable is good for everyone. The fear mongering from Anthropic and OpenAI looks like an attempt to corner the US market into using only US models so they can keep the profits, especially since China has proven that LLMs are a commodity. US AI labs should work on making LLMs cheaper or better harness. Altman and Dario are not trustworthy.
> "Anthropic believes that they are the ones who should have final say over how Anthropic is used; given that they think only they should be developing leading edge AI, they by extension think that only they should have final say over AI generally. When you further combine this realization with the company’s pronouncements about AI’s ability to conduct all economic activity, you realize that Anthropic’s leadership effectively wants to have power over everything and everyone."
> “My scorecard, at the highest level, would be we’ve been roughly right on technological predictions and pretty wrong on the social and economic implications” https://www.cxtoday.com/ai-automation-in-cx/sam-altman-softe...
I agree re: Dario quietly pushing for government control. He also said LLMs would replace a lot of entry-level information jobs, doubling the unemployment rate from 4-5% to 10%.
Yale did a study recently showing little impact on employment in high-AI exposed jobs https://budgetlab.yale.edu/research/ai-probably-not-yet-reas...
That's what Hollywood has been telling me my entire life!
In practice I'm not sure how big a deal this is - Bedrock is not the cheapest or best provider (try Fireworks AI or DeepInfra who do support more recent models), and would seem to be of more interest to corporate users who already have an AWS account.
Perhaps also of relevance the US government is hassling US companies that are using Chinese models (currently DoorDash, previously AirBnB & Cursor), which may limit demand for corporate use.
Edit to answer my own question:
License file [2] states:
> If the Licensee or any of its affiliates operates a Model as a Service business, and the aggregate revenue of the Licensee and its affiliates exceeds 20 million US dollars (or the equivalent in other currencies) in total over any consecutive 12 months, the Licensee must enter into a separate agreement with Moonshot AI before using the Software or its derivative works for any commercial purpose.
[1] https://huggingface.co/moonshotai/Kimi-K3
[2] https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE
Nice!
They don't explain how successful that went but it's a bit hilarious seen that an Anthropic dev explained that it's been 15 days Claude was hard at work --with nothing to show yet-- trying to rewrite itself in another language.
"You rewrite Claude Code, we rewrite oh-my-pi."
"You're nowhere after 15 days, we do it in 10."
Sure, it's apples to oranges and all that. But part of me thinks they know fully well what they did there.
> Qwen3.8-Max comes with the official support for reasoning_effort, which can be used to adjust reasoning depth and control cost:
> xhigh (default): for complex tasks demanding thorough analysis
> medium: balancing accuracy and speed
> low: efficient reasoning optimizing for speed and cost
I hope this is significantly cheaper. I've been loving Deepseek for it's nearly free usage costs, hard to justify switching from cents per day.
> Pricing: Input: $2.0 / M tokens Output: $6.0 / M tokens Implicit Caching: $0.25 / M tokens
The "Reproduce a research paper — then improve it" seems particularly impressive as well as actually useful - perhaps as close as we're going to get to "recursive self-improvement" given that these are data-driven not code-driven models.
I'm not sure how much real world use of AI is 100% hands off agentic coding, but I guess for objective evaluation purposes it needs to be something without a human in the loop.
It'd be interesting to see a comparison of each company's best models on a suite of real-world challenging tasks like this, but I guess difficult since each group of tasks would need to be one-time-use only to prevent subsequent benchmaxxing which makes comparisons useless.
Surely Alibaba is leading by example here by reducing work hours per week while keeping pay the same right? Right?
Ironically it's Electron BS, so actually, Sol could probably build itself...
However, they are not at the point where they can effectively train themselves, nor did they are capable of researching their own method of learning. SWEs in mid-corps on my country are right now relegated to reviews and sanity check, basically babysitting the LLMs and making sure they're not spouting nonsense. If you think about it, that's basically QA and can also be delegated to another AI. If Bun's rust rewrite that they tout as fully LLM-led can pass the test of time in a year or so I think that's it.
I believe all that is now constrained by compute and capital, not tech.
Already 38% of Americans aged 16 or older do not work and are not looking for work and yet are not dying from hunger or exposure. This would have seemed like a utopian dream to someone from the 1800s.
Being 16, 17 and 18 and not working is expected, given they are still at school. A good percentage of people aged 18 to 24 are studying full time.
This stat seems a bit misleading and doesn't support the "you can live and not work argument". How many of those people are retired? How many of those are under 18? How many of those simply cannot work? How many of those live at home or are on food stamp?
Are you not already experiencing this? I think this is fairly common for people using AI now, though the time may not always go into hobbies or sports. It's common for me to setup Claude with an hour+ task while I catch up on housework, or while I'm getting ready in the morning.
In the last couple of weeks I've unfortunately had multiple family illnesses - it has been helpful to have Claude keep up with much of my product development programming work while I visit my mother in hospital and check on my father's recovery. I'm able to give more time to family without worrying that business progress isn't keeping up. The overnight Claude sessions while I'm asleep have been particularly helpful.
It's infinite work, I just did more work while codex was doing it's thing in the background.
Lmao I love their video with the idea that people will be able to do their hobbies while ai does their job.
Before AI:
I have 10 tasks that take 1 hour each to do.
After AI:
I have 100 tasks that take 10 minutes each to do.
Same amount of time spent working, maybe even more stressful, just more productive.
And then you were doing 100 x 10 minutes = 100 / 6 hours = ~16 hours & 40 minutes.
Also I love how literal Qwen 3.5 was, hopefully 3.8 is still extremely literal. The token explosions were actually helpful in debugging prompts.
Their base models and architecture has quickly become the go-to for local inference and fine-tuning, even when they introduced some tricky things like GDN, so many people use it, that it was matter of days/weeks until lots of OSS frameworks adopted it.
There's a lot of AI models that each 'win' one week over the next. The pure definition of a commodity market and not a 'winner takes all market' as valuations would imply
Just like we have professors with specialisation wondering if AI models can also be so.
Intuitively it feels like focused models should be better models, but human programmers are ‘better’ knowing programming alongside general stuff. We’re not reduced by reading economics or Tolkien, removing such knowledge would be premature optimization.
If you throw general purpose model at a codebase - it will look at the table and data logical connections beyond what is explicitly declared. It will figure out on its own that Salaries should be displayed on SalariesTable.php and it will "know" that your prices should include vat and so on.
A human knows that VAT and price go together and are related, full size LLM does too, stripped one - doesn't.
>we have professors with specialisation
Yes, but any professor with a specialisation is an erudite with vast knowledge across the board.
I'd argue we can have a meaningful BIG model with a specialisation but not vice versa.
PS: all of this is about a model that is expected to be able to execute development tasks on a human level. Obviously we can have small models that are very capable in things like TTS or STT.
If this is true, this is a big deal for me, but unfortunately I cannot find anything in their legal agreements, so this marketing sentence is worthless.
Qwen doesn't overuse em-dashes, weird sentence structures with bold text: normal text.
In my private GitHub Repo Qwen finished as much tasks as Fable did. Without hitting 5-hour session limits. Qwen's token-per-second performance fluctuated greatly. From 20 tps up to 80 tps.
Qwen failed on some "reading between the lines", but so did Fable.
The only noticeable difference is skill loading. Until "Must use" in a skills front matter, Qwen tends to be very lightly loading skills. Superpowers works, because of the same Must use wording. All my local skills use the same wording now. For installed Skills, it's more or less my responsibility to remind Qwen using a certain skill.
I used Qwen together with Qwen Code. I didn't compare Qwen+Claude Code.
With my Owl code scanner, Claude and Qwen produced similar results. Qwen repeatedly used the same output format, despite no format provided and disabling memory. Fable varied greatly. Both followed the file format output and produced valid files.
With Qwen 3.8 being now a visual understanding model, I don't really miss Fable/Opus/Sonnet for my private projects.
At work we are still exclusively using Anthropic models with Claude Code. No change incoming on that front.
Privately, I'm sold on Qwen and Qwen Token plan. No session limits, many open weights models available via a single API.
Since it's horrifying thinking about what US companies do with my data, it's no worse or better thinking about what Chinese companies do. It's a choice between the Plague and Cholera.
You can always try out this model for free on Qwen Chat. Alibaba Cloud has too much friction for me.
> Qwen3.8-Max was asked to create the oh-my-cli project from scratch and, over a 10+ day long-horizon autonomous coding run
10+ days of building what exactly? Is that a shell prompt customization toolkit? Or a coding harness??
Neither - well, sort of the second. The poorly named thing is a self evolving coding harness, the self evolving part makes it a big deal.
I wish they had picked a different example.
How are you all justifying economical use of these local models right now? What's the cost efficient way to do this and do better (even with models evolving over time and losing now vs later) than the big labs?
The point of local models is privacy, offline use, and maybe no guard rails.
* Not talking about enterprises that buy DGX racks and host Chinese models for internal use.
But also, control and consistency. A local model cannot be changed out under your feet like an API model can be.
Sounds like we have similar boxes - mine has a 10 core CPU, 64 GB of ram, and a 2070 Super. My motherboard had two unused PCIe3x8 slots and doesn't support Blackwell GPUs. I bought a couple of brand new Ada generation RTX 2000s with 16GB of memory for under $1400 to get to 40 GB of VRAM. That will easily run Qwen3.6-27b at a 6-bit quantization and 80,000 token context size. It isn't fast (19-21 t/s), but using pi-coding-agent is fine.
Now, my instinct is that I am giving up SOTA performance on agentic coding with this setup and LLM. But the gap between my setup and SOTA commercial models is small enough that it doesn't matter to me.
We all know that is hugely subsidized, and I guarantee that OpenAI and Anthropic are looking to enshittify that ASAP.
The enterprise users, however, are not subsidized like that. They pay per token. And some developers in those companies are chewing down a lot of tokens. Self-hosting an open weight model could be a massive savings very quickly. It also gives them negotiation leverage when talking to OpenAI and Anthropic.
For all we know, inference might be dirt cheap, they might just be hiking the API prices so high for us to think subscriptions are subsidized.
The last thing I want my personal agent to do is to write up code and run arbitrary commands. That is practically a legalized RCE.
I mostly just need it to mostly be a very advanced NLP model that is able to figure the set of tools that it needs to call with what argument. Then it will just follow a predefined decision tree.
i might end up cancelling claude, anybody else thinking of the same ?
Uhh, sure... what could possibly go wrong?
It is clearly distilled with an aggressive pro West bias to increase sales in the West.
Does this mean they distilled Claude? Sounds like what Claude Code will often do.
What is important is how long it can go without requiring human intervention. Not just that it's possible to run on its own for a time.