Tri Dao

tridao

7

·

AI & ML interests

None yet

Recent Activity

authored a paper 1 day ago

FlashAttention-4: Algorithm and Kernel Pipelining Co-Design for Asymmetric Hardware Scaling

authored a paper about 1 year ago

M1: Towards Scalable Test-Time Compute with Mamba Reasoning Models

published an article over 1 year ago

Bamba: Inference-Efficient Hybrid Mamba2 Model

View all activity

Organizations

Articles 1

Article

62

Bamba: Inference-Efficient Hybrid Mamba2 Model

Papers 16

arxiv:2603.05451

arxiv:2504.10449

arxiv:2411.12372

arxiv:2408.15237

models 0

None public yet

datasets 0

None public yet