Event ·Official first-hand

NVIDIA 发布开源小模型 Nemotron 3.5 Bolt E4B 4GB，面向本地语音与机器人

NVIDIA 发布开源小语言模型 Nemotron 3.5 Bolt E4B 4GB，基于 Gemma 4 E4B 做后训练而成。该模型面向本地部署场景，目标是语音助手、游戏角色和机器人。显存需求约 4GB，可运行在 RTX 显卡和 Jetson 设备上。

Current read

NVIDIA 出了个 4GB 显存就能跑的开源小模型，从 Gemma 4 E4B 改的，能做本地语音助手和游戏 NPC，家里 RTX 或 Jetson 就能玩。

10 sources46° AI heatLast updated 2026/10/24 03:56:29

Evidence

10 sources

Primary sourcemarktechpost

2026/10/24 03:56:29View source

Related source 1Bindu Reddy

2026/10/23 04:00:41View source

Related source 2TestingCatalog

2026/10/23 07:55:08View source

Related source 3Merve Noyan

2026/10/22 06:54:10View source

Related source 4AI21 Labs

2026/10/22 14:47:04View source

Related source 5MiniMax_AI

2026/10/23 01:21:09View source

Related source 6kimmonismus

2026/10/23 18:35:00View source

Related source 7openclaw

2026/10/23 22:18:03View source

Related source 8Salesforce AI Research

2026/10/22 17:52:02View source

Related source 9VTV Công nghệ

2026/10/22 21:47:00View 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