How do the new Deepseek, Qwen, and Kimi models compare against GLM 5.3?
GLM-5.3: Frontier coding with emergent cyber capabilities
511–520 of 626 posts
Re: GLM-5.3: Frontier coding with emergent cyber capabilities
#512Earlier quoted context omitted.
What the person above is suggesting: * https://pi.dev/ * https://omp.sh/ (no personal opinions of either, links might be useful) I think that OpenCode is nice, their CLI version is enjoyable and their desktop/web version is okay : * https://opencode.ai/ I also quite like driving OpenCode through something like Kepler / Paseo and tools like that (with those I can still use my Anthropic Condition by Claude Code being t…
No one in their right mind would install software using `curl | bash`
I feel the same, which is why I only use these tools in a docker container.
Because life is compromise.
Re: GLM-5.3: Frontier coding with emergent cyber capabilities
#513Earlier quoted context omitted.
OpenAI and Anthropic are both seeking trillion IPOs, while Chinese labs are pumping out open-weight models that are free for US providers to host and monetize. These Chinese models cost less of US SOTA models to run, even if they are less capable. Providers can just run them, offer cheap tokens, and pocket the margin. I just don't see how you justify a trillion valuation for US AI labs when the underlying models are…
Another interesting potential market here will be 'LLM in a box'. All the hardware and other tooling in a prebuilt, but modular, package ready to go. Pay one up-front cost, get a system running [whatever open LLM] with a token rate of [x], optionally configured to be immediately ready for distributed usage. Basically the opposite of cloud stuff: no rent, no dependency, 100% guaranteed uptime, guaranteed security/priv…
so something like this? https://tinygrad.org/#tinybox
Re: GLM-5.3: Frontier coding with emergent cyber capabilities
#514Earlier quoted context omitted.
DwarfStar ( https://github.com/antirez/ds4 ) supports GLM 5.2 and DeepSeek. Not only for toying, but for getting work done.
Since GLM-5.3 has the same base model as 5.2, DwarfStar should support it as well, once the weights are released, right?
Nope
Re: GLM-5.3: Frontier coding with emergent cyber capabilities
#515Earlier quoted context omitted.
> I understand that such models can be used by malicious actors, but it’s fair to have it publicly available I feel like there should be some mechanism to prove you own the code/app/site/whatever and it will remove the guardrails from the LLMs allowing them to find and fix these vulnerabilities.
Isn’t this essentially what anthropic is doing, albeit in a manual fashion? They work with code owners to run mythos and find issues.
Code owners - no. Ultra wealthy code owners with connections, and a few peasants with popular projects, for public image.
Re: GLM-5.3: Frontier coding with emergent cyber capabilities
#516Earlier quoted context omitted.
Isn’t this essentially what anthropic is doing, albeit in a manual fashion? They work with code owners to run mythos and find issues.
OpenAI now makes it easy to join their verified security program. Took me 5 minutes, and I was able to get GPT to do a full end-to-end pen test
Is this about the new project Blue and Red thing? Or just pre-existing Trusted Access for Cyber program?
I've joined TAC, but still have to dance around it.
Re: GLM-5.3: Frontier coding with emergent cyber capabilities
#517Earlier quoted context omitted.
Don't we already do this with services like Let's Encrypt, which is arguably more sensitive? If you had the codebase you could fake it, but it would still provide some amount of protection against abuse.
With Let's Encrypt, all the verification is done on their side with them controlling the connection between themselves and whatever they're trying to verify. In this case, you can put whatever you want between the harness you're running (or modify the harness itself), and essentially "lie" to the model. Any verification technique would be fairly trivial to bypass, while you continue to run the harness locally.
Yeah, for web apps, you can trick models by simply proxying it and pointing the models to that localhost. They then think they're not working on a live target.
Have personally tested this with Opus and Sol and it works.
Classifiers are tricky though. Here's where open weights will win.
Re: GLM-5.3: Frontier coding with emergent cyber capabilities
#518Earlier quoted context omitted.
I’m in the Claude code harness for everything boat too. What are the alternatives?
I think that writing your own harness is a rite of passage now, just like writing your own search engine or database, rolling your own crypto… Anyways, please try mine! https://github.com/deepclause/deepclause-sdk
eventually we will get to “just use postrges” stage
Re: GLM-5.3: Frontier coding with emergent cyber capabilities
#519This is absolutely still shy of Sol and Fable, but only just by a hair. Ridiculous results. There's still not a compelling economic reason to drop OpenAI courtesy of the ludicrous reset addiction that's taken place, but it feels like we're on the precipice. How are you all toying with running this kind of thing in a mega quantized way locally? Two weeks out from released weights, but this is still just GLM 5.2 with p…
Each time I try to use GLM it is under heavy load and I get downgraded to the older model. So much so that I have given up trying to stop wasting my own time. I rather pay a few bucks more and not have to deal with that nonsense
Are you using their web frontend?
That seems really broken.
I also get downgraded there all the time, but via API all is fine.
Re: GLM-5.3: Frontier coding with emergent cyber capabilities
#520Earlier quoted context omitted.
You should try a better harness. Try pi, or ohmypi if you want a good OOB experience
what is a harness? The comments below are mixing IDE/ADE but other suggestions are purely terminal things and I don't get what their value is over just a terminal. Is a harness like a loop where it's just a vague thing that everyone nods about but everyone is nodding at something different?
The harness is the agent. LLM's can be asked to output things in JSON for example. The LLM then literally asks for things like "execute this cmd" or search/replace this string. The LLM outputs text, but in a deterministic format that can be parsed. The harness calls the LLM, exposes tools, executes tools the LLM asks for, gates tool use based on security controls. It's the runtime that the agent uses to do work.