Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Qwen3.8-Max: A New Bar for Coding and Cowork

qwen.ai

Post reply on HN
Print view View original

541–550 of 652 posts

Re: Qwen3.8-Max: A New Bar for Coding and Cowork

#541

Post by nearbuy »

tomjen3 wrote: post #522
ilaksh wrote: post #471

As someone who is searching for a new programming contract right now, reading all of the incredible abilities here is pretty intimidating. Especially since I get almost all of my projects from Upwork which is an outsourcing site. 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 as…

As someone else who is an IT consultant. 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…

The problem is Claude Fable is now better than most programmers I know at software architecture and performance optimization as well.

Re: Qwen3.8-Max: A New Bar for Coding and Cowork

#542

Post by julianlam »

wickedsight wrote: post #208
pettijohn wrote: post #37

Earlier quoted context omitted.

35B MoE is certainly a good and fast local model. I find 27B dense to be quite a bit smarter, so I daily drive that. I wish there was a ~100B MoE with maybe 10B active. It would be super smart and fast!

I use 27B in plan mode and 35B MoE in act mode. I noticed that is the best balance for me for consistent tool calls and intelligent planning. Takes some time to switch, but it's worth it for me.

Wonder if it's possible to share a common cache via cachyllama between 27B and 35B-A3B

Re: Qwen3.8-Max: A New Bar for Coding and Cowork

#543

Post by fr2029 »

hajile wrote: post #416
MontagFTB wrote: post #403

In a 30M+ LOC repo we blew out Qwen3.6’s context thinking about a code review. This was via direct call to model. Do harnesses facilitate better context management, or is there something else to accommodate its smaller context window?

The real answer is to stop trying to have AI reviewing AI code because it's just the blind leading the blind.

[dead]

Re: Qwen3.8-Max: A New Bar for Coding and Cowork

#544

Post by Forgeties79 »

enraged_camel wrote: post #348
OtomotO wrote: post #297

Earlier quoted context omitted.

> You at least need a capable machine, so that's not 0 monetary investment It is 0 monetary investment if I already have said machine lying around doing nothing. Which is exactly the story OP talked about.

But most people don't have an RTX 5090 lying around, so the story doesn't apply to them, right?

I'm running Qwen 27B no problem with an AMD 9070XT + 24gb DDR5 ram. Does basic web search for me (tool call with tavily, costs nothing I get 1000 searches a month) and is great for creative writing (primarily breaking writer's block). Until the recent surge in ram costs, that wouldn't be hard to do. I built the computer for ~$1600 a year ago.

Re: Qwen3.8-Max: A New Bar for Coding and Cowork

#545

Post by CamperBob2 »

akmarinov wrote: post #401
matheusmoreira wrote: post #169

Earlier quoted context omitted.

The NSA will run Mythos against the entire US population and it will find the patterns. Powerful computer purchases. Energy consumption suddenly spiking. HN comments demonstrating intent and/or contempt for the glorious self-evident freedom preserving laws of the United States. Then they call your local law enforcement with an anonymous tip about this suspicious dude who's long overdue for a totally random home inspe…

Yeah but I'm in Australia and just host AI models as a service - how are the NSA gonna convince the local law enforcement to do anything?

Ask Kim Schmitz. (Yes, I know, NZ is nowhere near Australia, the point still holds.)

Re: Qwen3.8-Max: A New Bar for Coding and Cowork

#546

Post by ilaksh »

tomjen3 wrote: post #522
ilaksh wrote: post #471

As someone who is searching for a new programming contract right now, reading all of the incredible abilities here is pretty intimidating. Especially since I get almost all of my projects from Upwork which is an outsourcing site. 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 as…

As someone else who is an IT consultant. 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…

That's a popular type of AI cope.

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.

Re: Qwen3.8-Max: A New Bar for Coding and Cowork

#547

Post by Bnjoroge »

badsectoracula wrote: post #231
throwaw12 wrote: post #227

Earlier quoted context omitted.

Qwen3.8-Max is the first in Qwen-Max series to be open-weight as well. 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

Kimi K3 is more like "weights available" in that you can download and use them but it is under a custom license that has a bunch of limitations where you have to pay Moonshot for doing some stuff. GLM 5.2 on the other hand is plain old MIT. Not sure how Qwen3.8-Max is going to be licensed, hopefully it'll be Apache like the smaller ones.

Kimi’s policy is extreme reasonable.

Re: Qwen3.8-Max: A New Bar for Coding and Cowork

#548

Post by ilaksh »

DANmode wrote: post #538
ilaksh wrote: post #471

As someone who is searching for a new programming contract right now, reading all of the incredible abilities here is pretty intimidating. Especially since I get almost all of my projects from Upwork which is an outsourcing site. 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 as…

I thought this comment was going to be about leveraging agents in ways your existing clientele demographic cannot or will not (yet|ever), but then you totally jumped the shark!

Can you give me a hint about wtf you are talking about?

Re: Qwen3.8-Max: A New Bar for Coding and Cowork

#549

Post by Bnjoroge »

qlte wrote: post #511
aleksandrm wrote: post #496

Earlier quoted context omitted.

US AI companies are already sweating and 100% pressuring the Trump administration for more anti-Chinese regulation, since there have already been talk of Trump considering banning Chinese models. There's however another push back from the startup industry urging them not to ban it, since it will stifle the innovation. In other recent news OpenAI also greatly cut their model prices, 20% for 5.6 Terra and 80% for 5.6 L…

> In other recent news OpenAI also greatly cut their model prices, 20% for 5.6 Terra and 80% for 5.6 Luna, to stay competitive. I’ve seen comments on HN saying how bad this is for the Chinese model developers since the cheaper option like Deepseek Flash are not longer as price competitive to justify the hassle/risk/lack of multimodal… but isn’t this a gigantic red flag for OpenAI/Anthropic at their current valuations…

Huh? V4Flash is still incredibly price competitive. The new version is right around GLM 5.2 and maybe slightly worse than opus 4.8 depending on which benchmark you use while being much cheaper(even factoring how most US zdr providers charge 10x Deepseek’s api caching price). K3 is also a tad behind fable/sol while being alot cheaper

Re: Qwen3.8-Max: A New Bar for Coding and Cowork

#550

Post by NekkoDroid »

Forgeties79 wrote: post #544
enraged_camel wrote: post #348

Earlier quoted context omitted.

But most people don't have an RTX 5090 lying around, so the story doesn't apply to them, right?

I'm running Qwen 27B no problem with an AMD 9070XT + 24gb DDR5 ram. Does basic web search for me (tool call with tavily, costs nothing I get 1000 searches a month) and is great for creative writing (primarily breaking writer's block). Until the recent surge in ram costs, that wouldn't be hard to do. I built the computer for ~$1600 a year ago.

I am getting ~13-15 tps with my 9070XT for the 27B (~35tps for the 35B-A3B), but I think for me the main bottleneck is the 64gb of DDR4 3600 memory. What kinda speeds are you getting with what speed of DDR5?
Post reply on HN