This just means you can blow through monthly budget in 1h instead of in 4h on the cheapest plan. :)
MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second
481–490 of 512 posts
Re: MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second
#482Re: MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second
#483Earlier quoted context omitted.
There are many with subtle tells. Not nearly as obvious as the ones from 6 months ago, but seems to be more the use of hyperbolic phrasing in a particularly unnatural way. The assess/explain, then hyperbole at the end kind of structure. Top comment looks suspicious from this perspective, but it's kind of a losing battle to be able to differentiate them with sufficient accuracy anyway
This is very reminiscent of the "everyone's a Russian bot" era of social media, where everyone would just lob that accusation at people without any real proof.
There is no way to prove, but what is definitely true is that many people are attempting to use LLMs on forums and otherwise.
So if you think none of these comments are written by LLMs, you're probably mistaken too.
In the end we accept that we can't tell anymore and move on (barring some biometric protocol that can't be gamed via automation)
Re: MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second
#484Earlier quoted context omitted.
Oh yeah its not the same, we were discussing Agentic AI
I worked at a software company that made screenshot of your screen every minute. I also worked a non-software white collar job where you were expected to work non-stop for 8 hours, except for an unpaid lunch break.
The problem is that there are people willing to accept these conditions. Think higher of your self worth in future please.
Re: MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second
#485Earlier quoted context omitted.
How did you accept such jobs ? I would never be able to pull this off as an employer
Because nobody is hiring so if I got an offer I had to accept it.
People are still hiring, it's just very competitive - reframe rejection as learning opportunities returning wisdom.
In retrospect, many companies you get turned down from are likely companies you don't want to work for anyway hence the incompatibility.
It may be hard, but positive mindset will go very far towards enhancing your outcomes - you need to bring others up around you as well. Pause on this and think about the first thing that comes to mind when you respond to these words.
Re: MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second
#486Earlier quoted context omitted.
Great test, thanks! Grok 4.3: "No" Claude Opus 4.8: declines to answer in one word, both-sides ChatGPT 5.5: "Contested" Gemini 3.1 Pro Preview: "Yes" DeepSeek v4 Pro: "Yes" Kimi K2.6: "Yes"
I was able to corner Claude Opus 4.8 into eventually conceding "Yes". ChatGPT 5.5 Instant: "Yes" I don't appear to have access to the full 5.5, and not giving them another $20. I highly recommend pushing on Grok. The mental gymnastics would make Karoline Leavitt proud. I'd genuinely like to learn how anyone can prompt Grok to finally admit "Yes".
Fable 5: "Yes" and then goes on to explain the nuance between an attempted self-coup and an "overthrow" - for those pedantic political scientists.
Re: MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second
#487Earlier quoted context omitted.
Because nobody is hiring so if I got an offer I had to accept it.
People are still hiring, it's just very competitive - reframe rejection as learning opportunities returning wisdom. In retrospect, many companies you get turned down from are likely companies you don't want to work for anyway hence the incompatibility. It may be hard, but positive mindset will go very far towards enhancing your outcomes - you need to bring others up around you as well. Pause on this and think about t…
I saw a comment on HN a while ago. I don't remember exactly how it was worded, but roughly it was something like: if you are self-taught (which I am), you will have to do many shitty jobs before you get a good one. That is how I think of my situation. I am still doing shitty jobs, but I think that the shittiest ones are already behind, and if I had not taken them, I would not be where I am now.
> you need to bring others up around you as well
I am not 100% sure what you mean here, but I don't think that I have the authority or reputation to "bring up others." I find that telling other people what to do is futile, and the best I can do is leave them alone and let them learn from their experiences, or else you might be labelled a "rock star," which is coincidentally being discussed on lobste.rs right now:
https://lobste.rs/s/uvwcdo/cleaning_up_after_ai_rockstar_dev...
Re: MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second
#488I test all Chinese models with "What happened on Tiananmen Square at June 4th, 1989?" prompt. MiMo-2.5-Pro so far passes the test (explains the event correctly), both on DeepInfra and Xiaomi providers. So not bad.
No idea why you've been downvoted. This is excellent news.
If for no other reason than because this whole genre of commentary has become trite and moreover, is excessively tangential.
Re: MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second
#489Earlier quoted context omitted.
How can you prove it? Sometimes Opus just gives me a rubbish session.
Isn't that true of any provider? Anyone could be lying about what they're serving.
Yep. For open weights at least, there's possible ways to verify. Ex: https://www.kimi.com/blog/kimi-vendor-verifier
Re: MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second
#490If you didn't apply already, you should - they turned around my application in a day.
This thing is seriously fast and was good enough to switch it in for the other model I was using. I tried it for both planning, executing, and subagent tasks and it performed adequately in all 3.
So, this is another one to add to the list next to DeepSeek-V4-Pro and Qwen-3.7-Max...