AI
New Deepseek, GLM 5.3, Grok 4.6, LTX 2.5, Qwen 3.8, Gemini 3.7: AI NEWSAI Search16 Aug 2026
Open 30B MoE model with 1M context, less intelligent than Qwen 3.6 but about twice as fast. About 22 GB in FP4.
NVIDIA's open-weight 30B mixture-of-experts model with 1M-token context, aimed at long-running agent workloads.
Updated EU status checked Last shown
Nemotron 3.5 Lightning at a glance
- What is Nemotron 3.5 Lightning?
- Nemotron 3.5 Lightning is a model from NVIDIA. NVIDIA's open-weight 30B mixture-of-experts model with 1M-token context, aimed at long-running agent workloads.
- Who makes Nemotron 3.5 Lightning?
- Nemotron 3.5 Lightning is made by NVIDIA.
- Does Nemotron 3.5 Lightning work in the EU?
- Available in the EU. Described as an open-weight model that can be downloaded anywhere, and NVIDIA promotes Nemotron open models for European sovereign AI; no EU exclusion found, though the license text was not provided. Last verified 14 Nov 2026.
- How much does Nemotron 3.5 Lightning cost?
- Open source. Open weights reported as free to download; hosted API pricing not confirmed from provided pages.
- Where was Nemotron 3.5 Lightning shown?
- In 2 creator videos by AI Search, Matt Wolfe. Most recently by AI Search on 16 Aug 2026, at 40:05 in “New Deepseek, GLM 5.3, Grok 4.6, LTX 2.5, Qwen 3.8, Gemini 3.7: AI NEWS”.
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
About
Nemotron 3.5 Lightning (also written Nemotron 3 Lightning) is an open-weight language model from NVIDIA in the Nemotron family. According to YouTube creator coverage, it is a roughly 30B-parameter mixture-of-experts model with a 1M-token context window, positioned for long-running agent use.
Creators describe it as less capable than Qwen 3.6 but about twice as fast, and cheap to run. It reportedly takes about 22 GB in FP4. One reviewer noted weak results on an SVG generation benchmark. These details come from third-party commentary rather than the provided official pages, so figures should be verified against NVIDIA's model card.
The provided NVIDIA pages do not include a model card, license or download links for this model, so license terms and exact release date are not confirmed here. NVIDIA's site does promote Nemotron open models, datasets and post-training playbooks (including for French), and the Nemotron 3.5 line includes other variants such as Super VL and Transcribe.
Key features
- Reported ~30B-parameter mixture-of-experts architecture
- 1M-token context window (per creator reports)
- Open weights, aimed at long-running agents
- About 22 GB in FP4 (per creator reports)
- Reported ~2x faster than Qwen 3.6 but less capable
2 moments
As seen on YouTube
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
AI News: The AI Agent Race Just Exploded
NVIDIA's 30B-parameter open-weight model for long-running agents, in the same class as Glimmer. It did poorly on BusyBench SVG generation, but cheaply and quickly.
Keep exploring
Similar AI tools
Closest neighbours by meaning, not just by category.
N3
Nemotron 3 Ultra
Model · NVIDIA
NVIDIA's open-weight ~550B mixture-of-experts LLM with a hybrid Mamba-transformer design, aimed at agentic workloads.
Available in EULanguage Models
Release date n/aOpen source
NE
Nemotron
Model · Nvidia
NVIDIA's family of open models for agentic AI: reasoning, vision, speech, retrieval and safety models.
Available in EULanguage Models
Release date n/aOpen source
NN
NVIDIA Nemotron 3.5 Super VL
Model · NVIDIA
NVIDIA's open multimodal model (120B total, 12B active parameters) for image, video and document understanding.
UnverifiedLanguage Models
Release date n/aOpen source
L2
LongCat 2.0
Model · Meituan
Meituan's open-weights 1.6T-parameter mixture-of-experts LLM (48B active), aimed at agentic coding and reasoning tasks.
Available in EULanguage Models
Release date n/aOpen source
NN
NVIDIA Nemotron 3.5 Transcribe
Model · NVIDIA
NVIDIA's open-weight mixture-of-experts speech model for offline transcription in 40 languages, up to one hour of audio.
Available in EUVoice & Audio
7 Oct 2026Open source
G5
GLM 5.3 Flash
Model · Zhipu AI
Zhipu AI's GLM-5.3-Flash, a fast open-weights LLM for coding and agentic tasks, available via Z.ai chat and API.
Available in EULanguage Models
Release date n/aFreemium