Unverified50% confidencePredictionExact time
The author said that if someone is willing to donate 8x H100 compute, he will try to use KS3 to go after the world record for nanogpt training speed, and claims that on proxy hardware he has already found a configuration that can beat the existing record.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/26/2026
First Seen
Valid until: 11/9/2026
Sources
Related Entities
Related Claims
UnverifiedVIDRAFT'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 0.9999.61% similarUnverifiedNVIDIA engineers, by writing a brand-new dedicated fused decoding kernel upstream for VLLM, achieved a throughput of 423 tokens per second per user on the 280-billion-parameter Kimi K3 model, a 1.68x speedup on equivalent hardware.59% similarUnverifiedH100/H200是英伟达专为大模型训练与推理设计的数据中心级GPU，采用NVLink高速互联技术使数千张卡协同工作58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1100455API
curl https://kongchang.com/api/v1/knowledge/claims/1100455MCP
get_claim(id=1100455)