Unverified50% confidenceOpinionExact time
VIDRAFT's release proves that backpropagation on an inference NPU is numerically credible, but the model has only 46M parameters, the training code is not open-sourced, and speed metrics were not disclosed; the engineering obstacles between 46M and larger-scale models currently cannot be judged.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/14/2026
First Seen
Valid until: 1/12/2027
Sources
Related Entities
Related Claims
Unverified'Slowing down' lacks an actionable definition, and no consensus has been reached on which metrics to slow (model parameter scale, training compute investment, or capability release schedules).72% similarUnverifiedPrecisely tracing the contribution of each segment of a large model's output back to its original training data is an engineering task that is nearly impossible to accomplish.70% similarUnverifiedThe 32k vocabulary collapsed at the 50M scale because its embedding matrix occupied nearly half of the model parameters and could not be adequately trained within 1,000 training steps.70% similarUnverifiedHumanEval has a limited sample size and has been extensively covered by training data, leading to suspicions of test set leakage in some models69% similarUnverifiedComparing a specifically trained 0.8B model against an untuned stock Qwen3.5-2B base model presents an issue of unequal comparison baselines.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1032403API
curl https://kongchang.com/api/v1/knowledge/claims/1032403MCP
get_claim(id=1032403)