@dMatrix
@dMatrix_AI
Infinite Inference. No Latency.
加入 February 2022
181 正在關注 1.5K 粉絲
845 tokens/sec per user. ~50% lower latency.
d-Matrix Corsair + NVIDIA H100 are redefining AI inference with attention-FFN disaggregation.
Faster inference. Smarter compute.
Read more:
顯示更多
