Releases·Model Release
NVIDIA releases Nemotron-SEA-LION-v4.8 30B-A3B and 120B-A12B models for Southeast Asian languages
A technical report introduces Nemotron-SEA-LION-v4.8, a family of Southeast Asian Languages in One Network models built on NVIDIA Nemotron 3, in 30B-A3B and 120B-A12B sizes with base and post-trained variants. Post-training with supervised fine-tuning and online on-policy distillation lifts SEA-HELM scores substantially across seven Southeast Asian languages. The release gives practitioners stronger regional-language models in efficient MoE configurations.
0 engagement1 sourceWed, Sep 16, 2026, 08:31 AM
AI Overview
- Nemotron-SEA-LION-v4.8, built on NVIDIA Nemotron 3, ships 30B-A3B and 120B-A12B base and post-trained variants
- On SEA-HELM, the 30B-A3B lifts the overall SEA score from 46.06 to 51.57; the 120B-A12B from 49.30 to 63.44
- Training uses Southeast Asian, reasoning, code, and multilingual parallel datasets, then SFT and online on-policy distillation
- Gains concentrate in instruction following, natural language reasoning, and understanding across seven Southeast Asian languages
Related
Releases·Model Release
NVIDIA Nemotron 3.5 Lightning now available in Amazon SageMaker JumpStart
NVIDIA has made Nemotron 3.5 Lightning, an open model optimized for high-volume agentic workloads, available on Amazon SageMaker JumpStart. The model is positioned as the fastest open model in its class for always-on agents, delivering up to 4x higher throughput. Users can deploy it without configuring serving infrastructure.
NVIDIA has made Nemotron 3.5 Lightning, an open model optimized for high-volume agentic workloads, available on Amazon SageMaker JumpStart. The model is positioned as the fastest open model in its class for always-on agents, delivering up to 4x higher throughput. Users can deploy it without configuring serving infrastructure.
0 engagement1 sourcerssMon, Aug 17, 2026, 06:06 PM
Releases·Model Release
NVIDIA releases Nemotron-3-Embed 1B and 8B text embedding models
NVIDIA released Nemotron-3-Embed, a collection of open embedding models led by an 8B model that ranks #1 on the RTEB leaderboard, along with efficient 1B variants. The models support multilingual retrieval across 34 languages and are designed for production-scale RAG, agentic retrieval, and code retrieval. Amazon SageMaker AI now offers serverless fine-tuning for these models.
NVIDIA released Nemotron-3-Embed, a collection of open embedding models led by an 8B model that ranks #1 on the RTEB leaderboard, along with efficient 1B variants. The models support multilingual retrieval across 34 languages and are designed for production-scale RAG, agentic retrieval, and code retrieval. Amazon SageMaker AI now offers serverless fine-tuning for these models.
0304 engagement3 sourcesreddit, rssThu, Jul 16, 2026, 12:00 AM
Releases·Model Releasefeatured
NVIDIA releases Nemotron 3.5 Lightning 30B A3B BF16 on Hugging Face
NVIDIA has released Nemotron 3.5 Lightning, a 30B-parameter mixture-of-experts model with 3B active parameters, designed for fast and efficient long-running agentic AI workloads. The release includes BF16 and NVFP4 quantized versions on Hugging Face, positioning it as the highest-efficiency model in its class for specialized tasks within multi-agent systems.
NVIDIA has released Nemotron 3.5 Lightning, a 30B-parameter mixture-of-experts model with 3B active parameters, designed for fast and efficient long-running agentic AI workloads. The release includes BF16 and NVFP4 quantized versions on Hugging Face, positioning it as the highest-efficiency model in its class for specialized tasks within multi-agent systems.
2301.1k engagement5 sourcesreddit, youtube, rss +1Tue, Aug 11, 2026, 01:19 PM
Releases·Model Release
llama.cpp PR adds support for NVIDIA Nemotron-3-Puzzle-75B-A9B, a 75.3B-parameter hybrid Mamba-MoE model
A llama.cpp pull request by YanissAmz adds support for NVIDIA's Nemotron-3-Puzzle-75B-A9B, a pruned-down hybrid Mamba-MoE-Attention model. The pruned variant shrinks the parent 120.7B-total-parameter model to 75.3B total / 9.3B active parameters, making it locally runnable today, though Multi-Token Prediction is not yet supported in llama.cpp.
A llama.cpp pull request by YanissAmz adds support for NVIDIA's Nemotron-3-Puzzle-75B-A9B, a pruned-down hybrid Mamba-MoE-Attention model. The pruned variant shrinks the parent 120.7B-total-parameter model to 75.3B total / 9.3B active parameters, making it locally runnable today, though Multi-Token Prediction is not yet supported in llama.cpp.
37 engagement1 sourceredditThu, Sep 3, 2026, 07:35 AM
Releases·Model Release
NVIDIA releases NemotronLabs VoiceChat-11B full-duplex model on Hugging Face
NVIDIA has released the NemotronLabs VoiceChat-11B model on Hugging Face, featuring full-duplex voice interaction. The model enables real-time, bidirectional conversation, allowing users to speak and listen simultaneously. This release advances voice AI capabilities for developers building conversational interfaces.
NVIDIA has released the NemotronLabs VoiceChat-11B model on Hugging Face, featuring full-duplex voice interaction. The model enables real-time, bidirectional conversation, allowing users to speak and listen simultaneously. This release advances voice AI capabilities for developers building conversational interfaces.
249 engagement1 sourceredditMon, Aug 3, 2026, 10:24 PM