Earlier quoted context omitted.
Just sign up for an AWS account and use the Anthropic models through Bedrock which Pi can use.
API costs are really high compared to subs.
Then you aren't the target market.
Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving
361–370 of 400 posts
Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving
#362Earlier quoted context omitted.
I have never seen a model be “lazy” before (I have seen them go for minimal change). I have been using the models through the api with various agents and no custom system prompt. So I am curious, how do people get these lazy outputs? Is it by having one of those custom system prompts that basically tells the model to be disrespectful? Or is it free tier? Cheap plans?
I have seen some people complain about a new tendency where it can suggest wrapping up the current task even though it isn't done yet. I haven't seen it myself though.
Usually this gets worse if you have a phrase like "wrap it up" earlier in the output, or if you're at a few hundred thousand tokens without compacting.
In both cases the fix is really simple, just compact.
Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving
#363Earlier quoted context omitted.
GLM 5.1 was the model that made me feel like the Chinese models had truly caught up. I cancelled my Claude Max subscription and genuinely have not missed it at all. Some people seem to agree and some don't, but I think that indicates we're just down to your specific domain and usage patterns rather than the SOTA models being objectively better like they clearly used to be.
It seems like people can't even agree which SOTA model is best at any given moment anymore, so yeah I think it's just subjective at this point.
They might be converging somewhat. The ultimate limiting factor is training data. Eventually I think they will converge and then the competition will be on memory and compute efficiency, with the best being the smallest maximally capable model.
Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving
#364Earlier quoted context omitted.
If this was true then they’d build services around those models and provide those for free or vastly cheaper than western competition. But that’s not what they’re doing. Instead they’re giving away the entire model for free. And by the way, Qwen isn’t build from some random entrepreneur who’s trying to solve the cold start problem, but from Alibaba which is a fucking behemoth. And surprisingly of course none of these…
> And by the way, Qwen isn’t build from some random entrepreneur who’s trying to solve the cold start problem, but from Alibaba which is a fucking behemoth. DeepSeek, Kimi, GLM, etc. are not built by behemoths, and they are free. You do not understand China's culture and market. > And surprisingly of course none of these models answer uncomfortable questions about China’s past. Download the GLM 5.1 weights and ask ab…
Excuse me if it’s considered uncouth on here to do this but, I would be interested in your thoughts on what I wrote here: https://news.ycombinator.com/item?id=47847600
I saw your comment after I wrote mine.
Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving
#365Earlier quoted context omitted.
rich people dont concern themselves with the cost of tokens.
It is not even rich. If you earn more than $30k it is worth your employer spending $3k on AI tools.
HN is very stingy for some reason, most people here make well over $150k and yet they are still worrying about $200/mo over $10/mo when the former will save more time.
Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving
#366Earlier quoted context omitted.
Ask a few Native Americans about dominance. Or maybe families of African descent. Or maybe families of Japanese Americans who lived in the US during WWII. Or maybe people of Latin descent living in the US today.
The US examples you just gave happened decades (and in some cases hundreds) of years ago. The difference is that it's happening in China right now, and nobody cares. You really don't see the difference?
People are dying in US detention camps right now.
https://www.motherjones.com/politics/2026/03/at-largest-ice-...
https://www.washingtonpost.com/business/2026/01/23/ice-detai...
Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving
#367Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving
#368Earlier quoted context omitted.
You're not giving an AI command line access to your work computer? How do you expect to keep up? /s
You give it command line access in a VM...
Yeah, fine... but it's like daily that a non-tech-savvy friend of mine tells me they just installed some shiny "harness" on their laptop now to organize their emails, and they "just put it in one folder" and "8n8 says", what does it say on the tin, Dave? "it says it's highly unlikely it will escape from the folder". Your work computer? "Yeah, but it's a real company. They're all about security."
So telling someone who just wants to upload an .xlsx file to a bot that they should just find a harness to give CLI access to their work computer - right after they say they work in a regulatory capacity - is just freakin malpractice.
Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving
#369Earlier quoted context omitted.
GLM 5.1 was the model that made me feel like the Chinese models had truly caught up. I cancelled my Claude Max subscription and genuinely have not missed it at all. Some people seem to agree and some don't, but I think that indicates we're just down to your specific domain and usage patterns rather than the SOTA models being objectively better like they clearly used to be.
I feel like it's Sonnet level for implementation, but not matching up to Opus for planning. But I agree it's close enough that it's worth using heavily. I've not cancelled my Claude Max subscription, but I've added a z.ai subscription...
My combo is codex and claude basic subscription for planing the hard tasks (if any) opencode with GLM 5.1 (z.ai coding plan) for the actual coding.
opencode is awesome I don't miss cluade or codex cli at all, and the z.ai plan is way more generous in compression.
I was lucky to subscribe to z.ai coding plan pro when it costed 30$/month, I was surprised now it costs 70$/month.
In case anyone wants to subscribe to z.ai with 10% discount [1] * here is the credit campaign rules * [2]
Re: Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving
#370Is a community LLM possible? We'd have code to dynamically construct the pre-training dataset and use P2P mechanisms to share the acquired dataset. It would involve peer-crawling and other mechanisms to allow many people to contribute chunks to the dataset. Crawling chunks would be dynamically allocated to those contributing to avoid any double-crawling.
For post-training, the dataset would be a bunch of code that orchestrates the creation of training data via LLMs (needs to be legally sound), plus some kind of mechanical turk approach (something like wikipedia, where volunteers can work on chunks of data).
The main mechanism is this: what is shared is not just code, but also the acquired training data.
Critical aspects: - to have a mechanism to peer-validate submissions to the data pool, so that everybody can donate data without the risk of vandalism - a mechanism where the weights go through distributed training stages; somehow devs should be able to get a "lock" on the weights, do a bit of post training on it, and then get it approved. The "lock" means that during this brief period (trainining run), other devs are informed so we don't get two set of branched weights. A mechanism auto-evals the weights and accepts them as the new, updated weights. Retroactive discarding of weighs (e.g. after revising evals) is possible by branching the weights (needs some kind of efficient deduplication to avoid many copies of the weights).
I think this is possible. Maybe not with RAM, GPU and power shortages though.
Main benefit: Trannsparent training set means you know what the model was trained for. This makes it less opaque and less trial-and-error to see what modality the model is good at. This helps harness builders but also any other users of the models. It also decentralizes power.