Prime Intellect Releases Prime Mega MoE: Trainable Fused MoE Kernel, 3.11x Faster Forward Pass
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.
In brief
- What: Prime Intellect Releases Prime Mega MoE: Trainable Fused MoE Kernel, 3.11x Faster Forward Pass
- When: October 29, 2026 at 21:28
- Source: MarkTechPost (media)
- Topic: Models
Prepared by Ilya Tarasov’s Tarasov Code newsroom based on a report by MarkTechPost.
