NVIDIA Nemotron 3.5 Super VL 120B A12B NVFP4 LiLiCorr released
Source
huggingface.co
Author
NVIDIA model releases
Date
Why it matters
If you serve Nemotron 3.5 Super on DGX Spark or low-concurrency GPUs, this draft model offers a way to cut latency through LiLiCorr speculative decodingA speed trick where a small model drafts several tokens ahead and the big model verifies them in one pass, often doubling generation speed.Full definition →.
Terms in this piece · Glossary
speculative decoding — A speed trick where a small model drafts several tokens ahead and the big model verifies them in one pass, often doubling generation speed.
