@Alibaba_Qwen @grok what is vllm?

@Alibaba_Qwen anyone know how much memory is needed to run this locally?

@Alibaba_Qwen can we please get a model with 300k native context instead?

@Alibaba_Qwen Is this another ROPE?

@Alibaba_Qwen So it can write a whole book all in one go?

@Alibaba_Qwen name is a bit short and easy to say don’t you think?

@Alibaba_Qwen Chatgp what? 🤣

@Alibaba_Qwen coming to qwen code or already there?

@Alibaba_Qwen Think AI was born polished? Read this. “Chatting with an Artificial Intelligence” shows the pre-hype texture of 2022: misses, leaps, and the first hints of voice. A rare primary source on how our conversations with machines began. #AI #ChatGPT https://www.amazon.com/dp/B0BP...

@Alibaba_Qwen open source strikes again. well done.

@Alibaba_Qwen You guys are just awesome. None of the AGI bs just practical stuff that actually works

@Alibaba_Qwen Wait, what? 🤯 Had no idea DCA even existed. Love these quick drop-ins. Now that’s what you call a truly open-sourced company.

@Alibaba_Qwen When are we going to get Qwen3 Omni series models?

@Alibaba_Qwen Boom! China is so gonna dominate the market in Open source - this is going to be awesome!

@Alibaba_Qwen @UnslothAI just got used to your model, guess Coder’s coming soon

@Alibaba_Qwen Damn china is rocking

@Alibaba_Qwen Are there any benchmark results for Qwen’s long context performance?

@Alibaba_Qwen Wow thanks! The east truly pushes open source.

@Alibaba_Qwen Awesome!

@Alibaba_Qwen 1M tokens? sheesh AIgust indeed.

@Alibaba_Qwen I think I will start looking at it

@Alibaba_Qwen This is amazing, thank you.

@Alibaba_Qwen magic ai labs do your thing.

@Alibaba_Qwen 1M token context?? guess i’ll start training for the ultramarathon too

@Alibaba_Qwen amazing.

@Alibaba_Qwen those guys just cant stop?

@Alibaba_Qwen Keep it coming. The same for coder?

@Alibaba_Qwen OpenAI in shambles

@Alibaba_Qwen qwen mogging again

@Alibaba_Qwen It just needs 1 TB RAM From the HF readme: To effectively process a 1 million token context, users will require approximately 1000 GB of total GPU memory. This accounts for model weights, KV-cache storage, and peak activation memory demands.

@Alibaba_Qwen what about the chat platform

@Alibaba_Qwen Great job, not like some gpt-ass

@Alibaba_Qwen 太好了，是 100 万上下文，番茄小说作者有救了！

@Alibaba_Qwen awesome wanna @Alibaba_Qwen to be the best at research also

@Alibaba_Qwen Amazing! 👏

@Alibaba_Qwen