This guy basically revolutionized local LLM. Please somebody fund him.
I'm getting ~50 tok/s on Qwen 3.8 Flash Next with 256k context using just an RTX 3090 + 48GB DDR4. That's insane.
Can't wait to see Qwen4 on this setup. No idea why more people aren't crazy about this.
People still dont fully grasp Strata
You can now buy old cheap DDR3 machine with 128GB or 64GB RAM connect 8GB+ GPU and run Qwen3.8-Flash-Next at 70tps+
Remember guys: More cores/threads in CPU - faster it will go :)
https://t.co/83diJSdDja
Nahhh bro, I'm gonna need some high-quality free tokens to keep developing my slop-ass anime-esque battle RPG 😭
Plz @QwenDevs , I need Qwen4 27B right nowwww
Qwen 4 — first results.
We tested the frontend and honestly it's incredibly powerful. It's right up there with Opus 5.5 and Astra, maybe just slightly behind, but it's way ahead of every other domestic beta.
Qwen 4 — first results.
We tested the frontend and honestly it's incredibly powerful. It's right up there with Opus 5.5 and Astra, maybe just slightly behind, but it's way ahead of every other domestic beta.
i can't stop crying. wtf was this episode. so tragic, i didn't feel this emotional reading the light novel but this episode was hard to watch. I wanna see this man-god dead right now, please. PLEASE !!! #無職転生#mushokutensei
Please @Alibaba_Qwen just drop the nuke. Can't wait for Qwen 4.0 with some next level ngram SSD offload shit. They're afraid. Open source must win. Fuck anthropic and openai
We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so.
Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.
You can read the full post here: https://t.co/OGyPb7yaYt