Unverified50% confidenceBenchmarkExact time
Meta wrote custom GPU code, Helion kernels, for the SGLang serving engine, achieving kernel speedups of 1.03x to 3.95x on Nvidia and AMD graphics cards, and a 3% end-to-end speedup on AMD MI350.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/15/2026
First Seen
Valid until: 1/13/2027
Sources
Related Entities
Related Claims
UnverifiedMoE GGUF models achieve up to 3x speedup on NVIDIA and AMD GPUs even when their size exceeds VRAM.72% similarUnverifiedMuse Spark Open ships with Helion GPU kernels integrated into SGLang, achieving 1.03–3.95× kernel speedups on NVIDIA and AMD GPUs.71% similarUnverifiedGPU版PyTorch需要英伟达GTX 6系列及以后的显卡，GPU处理速度通常是CPU的10-50倍69% similarUnverifiedAMD's RX 9700 series GPUs have hardware-level acceleration support for mxfp4, and upgrading to 128GB of memory and using the mxfp4 quantized version of Flash Next can increase QFN speed by approximately 2 times.68% similarVerifiedThe H100 GPU adopts NVIDIA's Hopper architecture, with a single-card peak compute of 3,958 TFLOPS (FP16 precision).67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1040255API
curl https://kongchang.com/api/v1/knowledge/claims/1040255MCP
get_claim(id=1040255)