Unverified50% confidenceFactExact time
Nemotron 3.5 Super VL uses a Mamba-2/Transformer hybrid MoE architecture, with 512 routed experts per MoE layer, paired on the vision side with a C-RADIO encoder, supporting text, JPEG/PNG images, and MP4 video up to 2 minutes long, with a context limit of 1M tokens and native support for multi-token prediction.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedNemotron 3.5 Super VL replaces a pure attention decoder with Mamba state-space blocks, uses a self-developed RADIO vision encoder, and natively supports NVFP4 four-bit quantization on the Blackwell architecture.76% similarUnverifiedNemotron-H 是 NVIDIA 于 2025 年发布的混合架构系列模型，将 Mamba 状态空间模型层与标准 Transformer 注意力层交替堆叠71% similarUnverifiedNemotron 3.5 Bolt's architecture adopts LatentMoE mixing Mamba-2 and Transformer, and the vision side uses the C-RADIO v4-H encoder.71% similarUnverifiedNemotron 3.5 Super VL uses a mixture-of-experts (MoE) architecture, routing each token to only about one-tenth of the expert networks.70% similarPartially VerifiedNVIDIA Nemotron 3.5 Super VL is a 120B-parameter MoE model with 12B active parameters, supporting up to a 1 million token context, aimed at document understanding, computer use and agent scenarios.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1049447API
curl https://kongchang.com/api/v1/knowledge/claims/1049447MCP
get_claim(id=1049447)