Unverified50% confidenceReleaseExact time
NVIDIA近期宣布为Vulkan提供端到端的描述符堆（Descriptor Heaps）支持，从API调用、驱动实现到硬件执行的整条链路都能原生处理描述符堆
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/7/2026
First Seen
Valid until: 10/5/2026
Sources
Vulkan描述符堆端到端支持：简化GPU资源绑定与Bindless渲染
rss6/25/2026
Related Claims
VerifiedNVIDIA从Volta架构开始引入Tensor Core，这是一种专门用于矩阵乘加运算的硬件单元72% similarUnverifiedNVIDIA近期为TensorRT引入了多设备推理支持（Multi-Device Inference Support）71% similarUnverifiedNVIDIA Hopper架构引入Confidential Computing功能，可在GPU内存中加密保护模型权重和推理数据71% similarUnverifiedNVIDIA近期为Dynamo框架新增了多轮智能体交互（Multi-Turn Agentic Harness）支持，将流式Token输出与工具调用做了深度集成71% similarUnverifiedNVIDIA Pascal架构（2016年）引入了线程级抢占（Thread-level Preemption）71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/185413API
curl https://kongchang.com/api/v1/knowledge/claims/185413MCP
get_claim(id=185413)