RedHat_AI releases optimized NVIDIA Nemotron 3.5 Super VL checkpoints

What happened

vLLM highlights @RedHat_AI’s FP8 and DSpark checkpoints for NVIDIA Nemotron 3.5 Super VL, a 120B multimodal MoE, using LLM Compressor and Speculators. RedHat_AI reports close to 2x performance on Hopper with accuracy intact. Its 4× H100 benchmarks show FP8 improving throughput under load and DSpark boosting per-user streaming speed, with both together outperforming either alone across HumanEval, math reasoning, and RAG workloads. Both checkpoints are open: https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-FP8 https://huggingface.co/RedHatAI/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B.dspark

The story so far

NVIDIA released Nemotron 3.5 Super VL and Nemotron 3.5 Bolt 74B-A9B 24GB, and both are now available on Ollama. A smaller Bolt E4B 4GB model is scheduled for Oct. 23. Super VL is an open multimodal model with 120 billion total parameters and 12 billion active parameters. It combines reasoning and tool use with text, image, document and video understanding for agent workflows. NVIDIA released its weights, datasets and training recipes. vLLM added launch-day support and lists a 1-million-token context window; Red Hat AI also released open FP8 and DSpark optimized checkpoints. Bolt 74B-A9B targets local agent tasks and uses native 2-bit quantization to fit in 24GB of memory. NVIDIA says it is optimized for RTX Spark and designed to run alongside other applications. It is also available on Hugging Face, with support for llama.cpp, LM Studio, Unsloth and vLLM.

Keep reading

Trending now

1. 1Meta Releases Muse Spark Open With Mac and Cloud Support, Topping US Open-Weight Index
2. 2Google Launches Gemini 4 Argon Globally, Expands Paid Access as Benchmark Results Vary
3. 3OpenAI Updates GPT-6 Astra Pro in ChatGPT With Parallel Agents and Faster Responses
4. 4Meta’s New Business Terms Raise Dispute Over Website Crawling and AI Training Rights