A

meme

[A @ababaka]

@0xSero Have you tried this model yourself? It thinks x5 from what qwen thinks! Completely unusable.

[A @ababaka]

@ornith_ Fingers crossed! 🤞 128gb ram bros waiting for good workhorse.

[A @ababaka]

@sudoingX Just backup plan when subscription limits are exhausted.

[A @ababaka]

@sudoingX Stepfun thinks a lot. I mean A LOT.

[A @ababaka]

@sudoingX Tested GPT 5.5 xhigh vs Qwen3.6 27b on same task. ChatGPT rated Qwen 7/10—consistent flaw: lacks code depth, misses key details. Tried varied system prompts, still always 7/10. So not 90 but 70%, still good

[A @ababaka]

@pupposandro The 1.5tb model is hard to call local.

[A @ababaka]

@stevibe Have you tried mimo m2.5 pro? It turned out to be unexpectedly good in my experiments. Plus, the 1 million context works well somewhere up to 300-400k.

[A @ababaka]

@ivanfioravanti Looks like q3 is total garbage.. look at metrics https://t.co/jp7AjYeLYF

[A @ababaka]

@SwedPaul Запихнут dgx spark в ноутбук. Скриньте.

[A @ababaka]

@jedisct1 @DgitalNarrative It's not the answer

[A @ababaka]

@stevibe is there at least one model worthy of attention besides the qwen3.6 27b, which successfully accommodates even 24gb of vram?

[A @ababaka]

@LottoLabs I wonder if there are agents who can parallel tasks in multiple threads well. In this case, 128GB of spark can be useful.

[A @ababaka]

@sudoingX but I can describe how hot it got in the room with the 3090 running.

[A @ababaka]

@ivanfioravanti Take a look at this. Probably you may add one more competitor https://t.co/M7x8kNLcYN

[A @ababaka]

@LottoLabs Where did the notes go? without run parameters, this site loses its meaning.

[A @ababaka]

@sudoingX Don't get your hopes up. in my experience and posts on reddit, with the quantization of qwen3.6 27b cache, model starts processing old messages in a loop. It took me a while to figure out what was going on, but that was probably it.

[A @ababaka]

Не прошло и 5 лет. Или прошло? https://t.co/iHiuHvXyLi

[A @ababaka]

@kadirnardev Qwen released 27b about month ago. What are you talking about?

[A @ababaka]

@UnslothAI Hmm... Can it really fit on 16gb vram with 200k context? Highly doubt