Unverified50% confidenceFactExact time
VKORE extends the NPU's executable operation set to cover the backpropagation primitives required for LoRA training, and is described as requiring no changes to NPU firmware.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/15/2026
First Seen
Valid until: 1/13/2027
Sources
Related Entities
Related Claims
UnverifiedVKORE's implementation approach is to recombine, in software, the training compute primitives missing from inference-only NPUs (such as gradient computation and weight updates) based on the low-level operations the NPU itself supports, and to maximize hardware utilization through scheduling.77% similarUnverifiedVKORE uses a software layer to fill the gap in NPUs' training capability, enabling NPUs to execute training workloads without changing their silicon design.76% similarUnverifiedVKORE's numerical behavior during training on NPUs is consistent with standard training; it is real training, not an approximation.71% similarUnverifiedThe instruction sets and data paths of inference-specific NPUs are often fixed and lack training primitives such as backpropagation and gradient reduction, which is the fundamental hardware limitation preventing them from training directly.67% similarUnverifiedThe core training technology of VKORE-46M-NPU is proprietary, is in the process of patent application, and details will be published in a subsequent paper; the core training code has not been open-sourced for the time being for intellectual property protection.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1035347API
curl https://kongchang.com/api/v1/knowledge/claims/1035347MCP
get_claim(id=1035347)