Know a great AI tool or AI news source? Submit it to AiByts
enterprise AI5 min read

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.

By Asif RazzaqMarkTechPost
Prime Intellect Releases Prime Mega MoE: Trainable Fused MoE Kernel, 3.11x Faster Forward Pass

Read the full story

Originally published by MarkTechPost on Oct 29, 2026.

Read on MarkTechPost(opens in a new tab)

Comments

Sign in to join the discussion.

No comments yet — be the first to share your thoughts.