# Prime Intellect Releases Prime Mega MoE: Trainable Fused MoE Kernel, 3.11x Faster Forward Pass

> Source: <https://www.marktechpost.com/2026/10/29/prime-intellect-prime-mega-moe-fused-moe-kernel-training/>
> Published: 2026-10-29 18:28:57+00:00

Prime Intellect released Prime Mega MoE, a backward pass for DeepSeek’s fused Mega MoE kernel. It makes the kernel trainable and speeds Qwen3 MoE training steps by up to 34% in prime-rl.

The post [Prime Intellect Releases Prime Mega MoE: Trainable Fused MoE Kernel, 3.11x Faster Forward Pass](https://www.marktechpost.com/2026/10/29/prime-intellect-prime-mega-moe-fused-moe-kernel-training/) appeared first on [MarkTechPost](https://www.marktechpost.com).