Event ·Multi-source

NVIDIA 发布 Nemotron 3.5 Super VL，长输入下仍保持高输出速度

NVIDIA 发布视觉语言模型 Nemotron 3.5 Super VL。该模型主打推理能力与输出速度的平衡，在输入达到 100K token 时仍能维持较高的输出速度。速度表现数据来自 Artificial Analysis 的对比测试。

Current read

NVIDIA 放出了 Nemotron 3.5 Super VL，是个视觉语言模型，亮点是输入拉到 100K token 还能保持输出速度，Artificial Analysis 有跑分对比。

11 sources22° AI heatLast updated 2026/10/15 15:09:13

Evidence

11 sources

Primary sourceNVIDIA AI

2026/10/15 15:09:13View source

Related source 1Artificial Analysis

2026/10/15 15:51:19View source

Related source 2TestingCatalog

2026/10/15 08:08:42View source

Related source 3vLLM

2026/10/15 15:50:21View source

Related source 4CoreWeave

2026/10/15 16:28:42View source

Related source 5Nous Research

2026/10/15 17:32:59View source

Related source 6marktechpost

2026/10/15 19:53:50View source

Related source 7Together AI

2026/10/15 15:01:33View source

Related source 8Scale AI

2026/10/15 15:12:57View source

Related source 9OpenRouter

2026/10/15 15:30:15View source

Related source 10Decoder

2026/10/15 16:09:44View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story