Unverified50% confidenceFactExact time
The official tweet did not disclose the model parameter count of Nemotron 3.5 Transcribe, the recognition accuracy for each language, the audio question-answering evaluation results, or the degree of quality loss at NVFP4 precision.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/29/2026
First Seen
Valid until: 11/12/2026
Sources
Related Entities
Related Claims
UnverifiedNo Decision 3.0 model was exposed to video data during training, yet on all 19,140 questions in the Perception Test validation set, the 27B version achieved a zero-shot score of 77.4 and the 4B version scored 70.3.63% similarUnverifiedThe source does not provide quantitative benchmark comparisons for VKORE-46M-NPU, including perplexity scores, downstream task accuracy, and training throughput in tokens per second.61% similarUnverifiedIn the Qwen3-4B comparative experiment, updating the model directly with unfiltered skills without using SGUID filtering decreased performance by 0.3 percentage points on HMMT25.60% similarUnverifiedPublic quantitative evaluation results are currently lacking on d3-ultra's multimodal dimension.60% similarUnverifiedThe team did not run the full-precision Qwen3.5 9B, so it cannot distinguish whether the performance drop stems from the model size being too small or the ternary quantization being too aggressive.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1110129API
curl https://kongchang.com/api/v1/knowledge/claims/1110129MCP
get_claim(id=1110129)