Unverified50% confidenceFactExact time
VIDRAFT pre-trained a 46-million-parameter language model, VKORE-46M-NPU, from scratch on a class of NPU hardware normally used only for inference, with the entire training process not relying on NVIDIA GPUs or the CUDA software stack.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
Partially VerifiedVIDRAFT released VKORE, an AI acceleration software layer for NPUs, which through software means gives NPUs originally capable only of inference the ability to train, without replacing hardware.79% similarUnverifiedGPUs have a large number of general-purpose compute cores and can be flexibly scheduled through programming models such as CUDA, making them naturally suited to the frequent matrix operations and gradient backpropagation in training; NPUs, by contrast, are dedicated chips customized for neural network inference, typically using systolic arrays or TPU-like dataflow architectures, highly optimized for fixed operators, but with closed programming models and limited support for dynamic computation graphs, making it traditionally difficult for NPUs to carry training workloads.77% similarUnverifiedVIDRAFT's VKORE software acceleration layer, without changing the chip, both increases NPU inference performance several times over and makes Transformer LoRA training run on this kind of inference-specific hardware, with numerical precision aligned with standard PyTorch at 0.9999.73% similarUnverifiedThe NPU built into Intel Core Ultra processors typically delivers 10-15 TOPS (INT8), which is suitable for AI inference but not for model training.72% similarUnverifiedVKORE's functional positioning is analogous to what CUDA is to NVIDIA GPUs, sitting between application-layer ML frameworks and NPU hardware, but the specific implementation details (whether it includes a compiler, kernel libraries, runtime scheduler) have not been disclosed.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1041047API
curl https://kongchang.com/api/v1/knowledge/claims/1041047MCP
get_claim(id=1041047)