# NVIDIA announces Nemotron 3.5 Super VL and two Bolt models for agent deployments [Canonical reading](https://escapemeridian.com/readings/cd7e5c49cbadb0299c2910afb97b30050c9791fa0bdad3ad89aab25d5eed8b78) NVIDIA says it has released Nemotron 3.5 Super VL, an open 120-billion-parameter model, alongside two Bolt models aimed at local and edge agent workloads. The company describes Super VL as fast and accurate for multimodal agent tasks, and says Bolt 74B-A9B can run locally within 24GB of memory. NVIDIA also says the models ship with open weights, datasets and training recipes, and can be customized for organizations’ own tasks. Performance and adoption statements here are NVIDIA’s claims. Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims. ## Original source NVIDIA · Article Original source date: Oct 15, 2026 (2026-10-15T15:00:24+00:00). Date basis: Date stated by the publisher. Added to Escape Meridian: 2026-10-16T01:07:40.976Z. First observed: 2026-10-15T15:12:34.659Z. Retrieved: 2026-10-16T01:07:10.533Z. - [Open From Data Center to Desktop: NVIDIA Expands Nemotron Family With Highest Efficiency Super VL and Bolt Models](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Limits This is an NVIDIA publisher announcement. Its performance comparisons, product descriptions and statements about customer adoption are not independently verified by the supplied evidence. The supplied text does not include detailed benchmark methodology for the speed and accuracy comparisons. ## Open question How much control does local deployment and customization give organizations over agent systems? ## Source-stated NVIDIA says it released Nemotron 3.5 Super VL, an open 120-billion-parameter model, and introduced two Bolt models for edge and local workloads. > NVIDIA today released Nemotron 3.5 Super VL, an open 120-billion-parameter model - [Supporting source](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Source-stated NVIDIA claims Super VL is five times faster than the next-best model in its class; the article does not independently establish this comparison. > As the fastest open agentic model in its class — 5x faster than the next best — - [Supporting source](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Source-stated NVIDIA says Bolt 74B-A9B uses native 2-bit quantization to fit in 24GB of memory and is designed for local agent tasks on specified NVIDIA hardware. > using a native 2-bit quantization to fit in 24GB of memory and run alongside other applications. - [Supporting source](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Source-stated NVIDIA says Super VL is distributed with open weights, datasets and training recipes, and can be adapted to organizations’ data and requirements. > The model ships with open weights, open datasets and training recipes. - [Supporting source](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Source-stated NVIDIA says organizations across several industries are post-training or customizing Super VL for specialized tasks and workflows. > Dozens of organizations across industries are post-training Super VL on their own data, tools, policies and workflows - [Supporting source](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Source and contributor context NVIDIA · Publisher The publisher is identified here. No individual author has been inferred. ## Source media Source-associated media was found; no additional asset was selected for this reading. The original source remains available. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## Benchmark A defined set of tests used to compare systems. Results depend on the tasks and measures selected. - [Definition reference](https://arxiv.org/abs/2211.09110) ## NVIDIA NVIDIA develops processors, networking and software used for AI, graphics and scientific computing. - [Explore this context](/entities/nvidia) - [Reference](https://blogs.nvidia.com/) ## Editorial provenance Version 1. Publication ID: 1980f1c3-d9bf-4fa0-8c8a-c4a51989604b. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: passed by anthropic/claude-opus-5.5 on 2026-10-16T01:07:40.976Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: material\_development. Observation: ceb122ff7165e408fa62ee144cff9a9c6866fc147e1e26cce48b3d98780bb9b6. Source text hash: 3d170edc4ecb919c7b84b41f7e4edbe9705160a7586da1a20799f2992ad26293. ## Revision and source-check history ### Version 1 — 2026-10-16T01:07:40.976Z NVIDIA says it has released Nemotron 3.5 Super VL, an open 120-billion-parameter model, alongside two Bolt models aimed at local and edge agent workloads. The company describes Super VL as fast and accurate for multimodal agent tasks, and says Bolt 74B-A9B can run locally within 24GB of memory. NVIDIA also says the models ship with open weights, datasets and training recipes, and can be customized for organizations’ own tasks. Performance and adoption statements here are NVIDIA’s claims. ## Original source NVIDIA · Article Original source date: Oct 15, 2026 (2026-10-15T15:00:24+00:00). Date basis: Date stated by the publisher. Added to Escape Meridian: 2026-10-16T01:07:40.976Z. First observed: 2026-10-15T15:12:34.659Z. Retrieved: 2026-10-16T01:07:10.533Z. - [Open From Data Center to Desktop: NVIDIA Expands Nemotron Family With Highest Efficiency Super VL and Bolt Models](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Limits This is an NVIDIA publisher announcement. Its performance comparisons, product descriptions and statements about customer adoption are not independently verified by the supplied evidence. The supplied text does not include detailed benchmark methodology for the speed and accuracy comparisons. ## Open question How much control does local deployment and customization give organizations over agent systems? ## Source-stated NVIDIA says it released Nemotron 3.5 Super VL, an open 120-billion-parameter model, and introduced two Bolt models for edge and local workloads. > NVIDIA today released Nemotron 3.5 Super VL, an open 120-billion-parameter model - [Supporting source](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Source-stated NVIDIA claims Super VL is five times faster than the next-best model in its class; the article does not independently establish this comparison. > As the fastest open agentic model in its class — 5x faster than the next best — - [Supporting source](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Source-stated NVIDIA says Bolt 74B-A9B uses native 2-bit quantization to fit in 24GB of memory and is designed for local agent tasks on specified NVIDIA hardware. > using a native 2-bit quantization to fit in 24GB of memory and run alongside other applications. - [Supporting source](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Source-stated NVIDIA says Super VL is distributed with open weights, datasets and training recipes, and can be adapted to organizations’ data and requirements. > The model ships with open weights, open datasets and training recipes. - [Supporting source](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Source-stated NVIDIA says organizations across several industries are post-training or customizing Super VL for specialized tasks and workflows. > Dozens of organizations across industries are post-training Super VL on their own data, tools, policies and workflows - [Supporting source](https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/) ## Source and contributor context NVIDIA · Publisher The publisher is identified here. No individual author has been inferred. ## Source media Source-associated media was found; no additional asset was selected for this reading. The original source remains available. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## Benchmark A defined set of tests used to compare systems. Results depend on the tasks and measures selected. - [Definition reference](https://arxiv.org/abs/2211.09110) ## NVIDIA NVIDIA develops processors, networking and software used for AI, graphics and scientific computing. - [Explore this context](/entities/nvidia) - [Reference](https://blogs.nvidia.com/) ## Editorial provenance Version 1. Publication ID: 1980f1c3-d9bf-4fa0-8c8a-c4a51989604b. Prepared by gpt-6-luna; grounding review by gpt-6-luna. Source check: passed by anthropic/claude-opus-5.5 on 2026-10-16T01:07:40.976Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: material\_development. Observation: ceb122ff7165e408fa62ee144cff9a9c6866fc147e1e26cce48b3d98780bb9b6. Source text hash: 3d170edc4ecb919c7b84b41f7e4edbe9705160a7586da1a20799f2992ad26293. ### passed — 2026-10-16T01:07:40.976Z — anthropic/claude-opus-5.5 Publication: 1980f1c3-d9bf-4fa0-8c8a-c4a51989604b attribution — pass: The draft attributes the claims to NVIDIA, which is the publisher of this announcement. Performance and adoption statements are framed as NVIDIA's claims. > NVIDIA today released Nemotron 3.5 Super VL, an open 120-billion-parameter model summary — pass: The title, summary and question match the source. The source announces Super VL and two Bolt edge models, and the summary reflects its emphasis on local deployment, open weights and customization. The question about control follows from the source's section on control and openness. > Joining it are two new edge models: Nemotron 3.5 Bolt 74B-A9B 24GB date — pass: The originalPublished value (2026-10-15T15:00:24Z) comes from the publisher page and matches the feed entry. The source's related-news listing also dates the article Oct 15, 2026. Nothing in the evidence indicates an earlier first publication. > Open From Data Center to Desktop: NVIDIA Expands Nemotron Family With Highest Efficiency Super VL and Bolt Models Oct 15, 2026 support — pass: Each claim's excerpt is an exact quote from the source and supports the attributed statement. These are the 120B release, the 5x speed claim, the 2-bit quantization fitting 24GB, the open weights, datasets and recipes, and adoption by dozens of organizations. The draft makes no independent-verification overreach. > As the fastest open agentic model in its class — 5x faster than the next best — media — pass: The selected media is the verified direct link to the original NVIDIA article, which is the most direct way for readers to access the source. ## Corrections and challenges [Send a correction or an opposing source](https://escapemeridian.com/corrections/) [JSON record](https://escapemeridian.com/readings/cd7e5c49cbadb0299c2910afb97b30050c9791fa0bdad3ad89aab25d5eed8b78.json)