Mastodawn

🚀 Ah yes, the latest tech buzzword salad! Mamba-3 is apparently speeding things up faster than it takes to blink, and now we can all rent fancy #NVIDIA #GPUs as if we're borrowing sugar from a neighbor. 🍬 Who knew "accelerated compute" was just a synonym for "we added a few more buzzwords and called it innovation"? 😂
https://www.together.ai/blog/mamba-3 #techbuzzwords #acceleratedcompute #innovation #Mamba3 #HackerNews #ngated

Mamba-3

Meet Mamba-3: the SSM built for inference. Faster than Transformers at decode, stronger than Mamba-2, and open-source from day one.

Hacker News Mar 21

Mamba-3

https://www.together.ai/blog/mamba-3

#HackerNews #Mamba3 #AI #Innovation #Technology #TogetherAI

Mamba-3

Meet Mamba-3: the SSM built for inference. Faster than Transformers at decode, stronger than Mamba-2, and open-source from day one.

sayzard Mar 18

Sebastian Raschka (@rasbt)

Mamba-3이 출시되었으며, 작성자는 Mamba 및 유사 모델들이 트랜스포머 어텐션 하이브리드 아키텍처(Qwen3.5, Kimi Linear 등)에서 흥미로운 활용처라고 평가합니다. 다음 세대 하이브리드에서 Gated DeltaNet 대신 RoPE가 추가된 Mamba-3을 교체해보는 실험을 제안하고 있습니다.

https://x.com/rasbt/status/2034088726997893168

#mamba3 #transformer #qwen3.5 #gateddeltanet #rope

Sebastian Raschka (@rasbt) on X

Oh wow, Mamba-3 is here! For me, the most interesting use case of Mamba and Mamba-likes are the recent transformer attention hybrid architectures (Qwen3.5, Kimi Linear, etc.) Would be interesting to swap Gated DeltaNet with Mamba-3 (which now also has RoPE) in next gen hybrids.

X (formerly Twitter)