Linoy TsabanX (Twitter)My favorite editing model, FLUX.2 [klein] 9B, just got 2x faster:打开视频页面383在 X 上点赞分享X在 X (Twitter) 上打开
Haocheng XiX (Twitter)🎥 Video generation is hitting the memory wall.打开视频页面270在 X 上点赞分享X在 X (Twitter) 上打开
Fireworks AIX (Twitter)Fireworks blazing fast LLM inference is now available on Poe!打开视频页面191在 X 上点赞分享X在 X (Twitter) 上打开
TheLAPurchaserX (Twitter)KV cache yada turboquant yada yada compression something something打开视频页面70在 X 上点赞分享X在 X (Twitter) 上打开
EXO LabsX (Twitter)But the KV cache is created for each transformer layer.打开视频页面255在 X 上点赞分享X在 X (Twitter) 上打开
Pavlo MolchanovX (Twitter)🚀 Self-speculation brings 6.75x real speedup for LLM generation with SGLang inference!打开视频页面579在 X 上点赞分享X在 X (Twitter) 上打开
Vikas guptaX (Twitter)🚨 YOUR GPU IS PROBABLY WASTING MORE THAN YOU THINK.打开视频页面204在 X 上点赞分享X在 X (Twitter) 上打开
Marc SunX (Twitter)`transformers` + `torchao` quantization + `torch.compile` for faster inference speed and less memory…打开视频页面156在 X 上点赞分享X在 X (Twitter) 上打开
SakataX (Twitter)UNCENSOR ANY AI MODEL ON YOUR OWN GAMING PC打开视频页面敏感内容此视频可能包含敏感内容。查看设置716在 X 上点赞分享X在 X (Twitter) 上打开
Tech with MakX (Twitter)These are literally the best visual resources for learning how Transformers, LLMs, embeddings…打开视频页面1225在 X 上点赞分享X在 X (Twitter) 上打开
Aleph AlphaX (Twitter)We are excited to launch our two models Pharia-1-LLM-7B-control and Pharia-1-LLM-7B-control-aligned.打开视频页面192在 X 上点赞分享X在 X (Twitter) 上打开
Ettore Di GiacintoX (Twitter)Depth Anything 3 now runs as pure C++/ggml (ggml) .打开视频页面626在 X 上点赞分享X在 X (Twitter) 上打开
Harsha GundalaX (Twitter)They were building in stealth for 2 years, I was building in stealth for 2 hours…打开视频页面5791在 X 上点赞分享X在 X (Twitter) 上打开
Ilir AliuX (Twitter)Robots can now reconstruct 3D scenes in real time from a single RGB camera.打开视频页面853在 X 上点赞分享X在 X (Twitter) 上打开
Horace HeX (Twitter)Happy to OSS gpt-fast, a fast and hackable implementation of transformer inference in <1000 lines of…打开视频页面2235在 X 上点赞分享X在 X (Twitter) 上打开
RuliadX (Twitter)Introducing DeepThought-8B: Transparent reasoning model built on LLaMA-3.1 with test-time compute…打开视频页面1005在 X 上点赞分享X在 X (Twitter) 上打开
Daniel SanX (Twitter)Absolutely recommend using Cache Control in Claude Code打开视频页面2771在 X 上点赞分享X在 X (Twitter) 上打开
KKaruparti AITikTokComment “AI” and I’ll share the complete 9-week roadmap to become a job-ready AI Engineer.打开视频页面364在 TikTok 上点赞分享TikTok在 TikTok 上打开