Unverified50% confidenceFactExact time
Version 5.18.0 unified device_mesh initialization to support pipeline-parallel plus tensor-parallel inference, and combined FSDP2 with expert parallelism via a two-dimensional (fsdp, tp) device mesh.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/4/2026
First Seen
Valid until: 1/2/2027
Sources
Related Entities
Related Claims
UnverifiedUnsloth's save_pretrained_merged now supports compressed FP8/FP4 export, and the Studio export UI has added imatrix GGUF and compressed export options.62% similarUnverifiedNVFP4是NVIDIA私有的FP4实现，MXFP8是微软MX浮点标准的FP8变体，二者为Blackwell架构专属格式61% similarUnverifiedFSDP is the officially released fully sharded data parallelism solution from PyTorch, which reduces per-device memory usage from O(N) to O(N/world_size) by sharding model parameters, gradients, and optimizer states across different devices.60% similarVerifiedNVFP4 is a 4-bit floating-point (FP4) quantization format introduced by NVIDIA, designed specifically for the hardware acceleration capabilities of Blackwell architecture GPUs60% similarUnverifiedglTF格式由Khronos Group于2017年制定，被称为「3D界的JPEG」，几何数据以二进制缓冲区存储，材质采用PBR工作流60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/972013API
curl https://kongchang.com/api/v1/knowledge/claims/972013MCP
get_claim(id=972013)