[flagged]
DeepSeek V4 – almost on the frontier
101–110 of 420 posts
Re: DeepSeek V4 – almost on the frontier
#102I tried deepseek v4 through open code at the weekend. I'm a daily Claude/Claude code user. I tried to build something simple and while it got the job done the thinking displayed did not fill me with confidence. It was pages and pages of "actually no", "hang on", "wait that makes no sense". It was like the model was having a breakdown. Bear in mind open code was also new to me so I could be just seeing thinking where…
use hide_thinking in opencode to get the claude experience :p
Re: DeepSeek V4 – almost on the frontier
#103Has anybody used V4 hard, for the most challenging tasks (agentically, locally)? It's so hard to compare without putting serious time in it. Like spending a year daily with the model.
I tried it for two tasks using Claude Code, on max effort. 1. Web platform, asking it to analyse a feature to create reports, and coming up with better solution and better UX. it did great, I would say on par with Sonnet 4.6 or even opus considering the thinking and explanation 2. Mac app with some basic functionality, it did well from functional perspective but then I used Opus 4.7 to evaluate and suggest improvemen…
Claude Code poisons non-anthropic models in usage. We found this out when the code was leaked. Use a fork or OpenCode/pi-coding-agent
Re: DeepSeek V4 – almost on the frontier
#104I'm not sure I'd call it "almost on the frontier," but I do think that v4 Pro is the most usable coding model I've seen out of China. I've used it via Ollama Cloud (coding) and OpenRouter (data processing). Feels Sonnet-level to me -- solid at implementation when given a specification, but falls a good bit short of Opus 4.7 max thinking when planning out larger changes or when given open-ended prompts.
Have you given GLM 5.1 or Kimi K2.6 a shot for coding? They outperform Deepseek v4 pro.
Glm5.1 is fantastic for me. But that could be how I use it, I don't ask it to build entire apps or entire features, instead asking it to build piecemeal functionality. For that it compares very well to chatgpt 5.4 (I haven't extensively tried 5.5, it might be better, might be same). I have given deepseekv4 pro a try but not much more than a try, as it performed subpar on 4 tasks in a row (missing the obvious/intended path, generating subpar slightly buggy code to make things work the not obvious way) , I gave up on it.
Glm5.1 for me was a bit of a llama3.1 moment (first open model i could chat with that was usable in manging my inputs the intended way) for code, the first open model that was actually usable.
Re: DeepSeek V4 – almost on the frontier
#105The pelican is really getting old as an a standalone evaluation metric. By now they are certainly going to be in training set if not explicitly tuned to produce it for the press on HN alone. Keep the pelican but isn’t it time to add something else more novel that all current and past models struggle with?
One shot canvas and svg images or animations are also just something that at this scale shouldn't be an issue at all, even Qwen running locally on 24gb cards can do impressive ones.
Don't understand why this test gets any attention, I mean other than the pelicans which isn't a good test, theres no meat in this article.
Re: DeepSeek V4 – almost on the frontier
#106I tried deepseek v4 through open code at the weekend. I'm a daily Claude/Claude code user. I tried to build something simple and while it got the job done the thinking displayed did not fill me with confidence. It was pages and pages of "actually no", "hang on", "wait that makes no sense". It was like the model was having a breakdown. Bear in mind open code was also new to me so I could be just seeing thinking where…
> "actually no", "hang on", "wait that makes no sense" Claude does the same thing, claude code just hides the thinking now
And before that they summarized it. But yeah, thinking was always like that (when it first started, it almost just seemed like a scheme to massively increase token use..)
Re: DeepSeek V4 – almost on the frontier
#107This gives me hope that when the subsidization circus ends and everyone is on pure usage then it won't be entirely exclusionary to mere mortals who don't have $200pm budgets.
IMO there are two things that make me optimistic that we won’t see a big rug pull where price-to-capability ratio skyrockets relative to today: * As you’ve noted, people keep finding ways of slamming more intelligence into smaller models, meaning that a given hardware spec delivers more model capability over time. * Hardware will continue to improve and supply will catch up to demand, meaning that a dollar will deliv…
I also hope that we’ll find effective ways to distribute load between small local models and heavyweight remote models. Sort of like what Apple tried to do in iOS.
So much of what I ask codex to do doesn’t require full GPT 5 intelligence, and if 75% of the tokens were generated locally that’d save a massive amount of cost.
Re: DeepSeek V4 – almost on the frontier
#108This gives me hope that when the subsidization circus ends and everyone is on pure usage then it won't be entirely exclusionary to mere mortals who don't have $200pm budgets.
By the time the dust settles I wouldn't be surprised if personal interactive usage couldn't even be had for under $200. I can't fit my modelling of the serving costs of these things to any public reporting, even the more bearish examples
Re: DeepSeek V4 – almost on the frontier
#109Earlier quoted context omitted.
I tried it for two tasks using Claude Code, on max effort. 1. Web platform, asking it to analyse a feature to create reports, and coming up with better solution and better UX. it did great, I would say on par with Sonnet 4.6 or even opus considering the thinking and explanation 2. Mac app with some basic functionality, it did well from functional perspective but then I used Opus 4.7 to evaluate and suggest improvemen…
Claude Code poisons non-anthropic models in usage. We found this out when the code was leaked. Use a fork or OpenCode/pi-coding-agent
By poisons, do you mean it degrades their quality of output somehow?
Re: DeepSeek V4 – almost on the frontier
#110Earlier quoted context omitted.
As opposed to? Do you really think OpenAI, Anthropic or any other entity in the same business respects your data? The Chinese AI companies who release open weights actually deserve whatever input you give them. They are the reason why there is competition and not duopolies in the domain.
I think Google, and likely Anthropic, indeed do honor the settings chosen by the user. For Google in particular it'd be very surprising if they didn't. That's also why both do everything they can to trick users into allowing it. OpenAI, I wouldn't be surprised if you were right.
The models doesn’t get better by themselves. You’re naive.