# Prime Intellect releases Prime Mega MoE, a trainable fused MoE kernel

> Original title: Prime Intellect Releases Prime Mega MoE: Trainable Fused MoE Kernel, 3.11x Faster Forward Pass

- Source: MarkTechPost
- Published: 2026-10-29T18:28:57.000Z
- Bad Influence: https://badinfluence.news/items/gxr4eqx2pd44je6mauhmjxy7y
- Original: https://www.marktechpost.com/2026/10/29/prime-intellect-prime-mega-moe-fused-moe-kernel-training/

## Summary

Prime Intellect has released Prime Mega MoE, a training extension of DeepSeek's Mega MoE kernel that adds a fused, overlapping backward pass. Self-reported tests on Qwen3 MoE models show full training steps 32% to 34% faster than the prime-rl baseline and 48% to 55% faster than TorchTitan, with the forward pass up to 3.11x faster on a 30B layer. The kernel supports BF16, MXFP8 on Blackwell, and blockwise FP8 on Hopper, and ships in prime-rl as the mega_moe dispatch backend under Apache 2.0.