{"resourceId":"nvidia-nemotron-35-super-vl-blog-20261015","versions":[{"version":"external-4677bed6a1f33af14140bfab0e0bd48b614df60fd86cd4d7b174f5b185f21b93","resource":{"id":"nvidia-nemotron-35-super-vl-blog-20261015","title":"NVIDIA releases open Nemotron 3.5 Super VL and Bolt models for agents from data center to desktop","organization":"NVIDIA","sector":"Technology vendor / AI platforms","geography":"Global (U.S. vendor; Hugging Face / NIM distribution)","publishedAt":"2026-10-15","publicationDate":"2026-10-15","eventDate":"2026-10-15","sourceName":"NVIDIA Blog","sourceLabel":"Vendor product announcement","sourceUrl":"https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["developers-agents","infrastructure","knowledge-work","data-security"],"finding":"On Oct 15 NVIDIA released Nemotron 3.5 Super VL, an open 120-billion-parameter multimodal model for specialized always-on agents, plus Nemotron 3.5 Bolt 74B-A9B 24GB for local agentic workloads; Bolt E4B 4GB is due Oct 23. Super VL is positioned for instruction following, tool use, multistep workflows and visual understanding of PDFs, charts and webpages, with claimed class-leading speed (up to 5x). Organizations can post-train on their own data and deploy on their own infrastructure. Named customizers include Scale AI (Skydance, EY), Ramp, AI21/Alice, Mercor and others; Salesforce is said to be customizing Nemotron for AI agents that digitize and automate complex government workflows via Missionforce Operations. Distribution: Hugging Face, OpenRouter, ModelScope and NVIDIA NIM; NeMo for customization; Dynamo claimed to serve up to 8x more users with 85% lower latency.","sledRelevance":"Direct platform signal for SLED buyers evaluating open multimodal agents that can stay on-premises; Salesforce Missionforce government-workflow mention is a vendor claim needing buyer validation, not a measured SLED deployment.","evidence":"Primary vendor blog dated Oct 15, 2026 by Kari Briski; partner results are vendor-attributed; no independent measurement in this page.","architectureImplications":"Open weights plus optional NIM packaging; runs across DGX Spark/Station, RTX PRO/RTX, Jetson, data center and cloud; Bolt fits 24GB local memory with native 2-bit quantization; NeMo for post-training.","governanceImplications":"Vendor emphasizes customer control of post-training data, tools and policies; production governance still buyer-owned.","securityPrivacyImplications":"On-premises deployment framed as keeping data private; NIM said to refresh security patches every two weeks—buyers must still assess residency, access control and model-change governance.","caveats":"Vendor claims for 5x speed, Dynamo 8x/85% and partner outcomes are not independently verified here; government workflow mention is Salesforce customization intent, not a public-sector case study with metrics.","streamIds":["nvidia"],"roles":{"sales":"Sales framing: NVIDIA's Oct 15 Nemotron 3.5 Super VL and Bolt release offers open multimodal agents that agencies can post-train and run on their own stack, including a Salesforce Missionforce claim about automating complex government workflows. Interpretation: use this as a discovery opener for document-heavy knowledge work and local agents, not as proof of SLED outcomes. Bound the conversation to evaluation of open weights first, then production packaging and license cost, and treat every named partner gain as vendor-attributed until the buyer reproduces it.","engineering":"Engineering framing: Super VL is pitched as a single multimodal agent model for PDFs, charts, webpages, tool use and multistep workflows, with Bolt for 24GB local systems and optional NIM plus Dynamo serving. Pre-sales engineers should map candidate workloads to on-prem versus NIM versus open-framework serving, confirm GPU memory and quantization paths, and separate open-weight evaluation from any AI Enterprise-supported production design. Salesforce Missionforce is an integrator claim—validate government workflow scope and data boundaries before architecture lock-in.","delivery":"Delivery framing: standing up Super VL means choosing open-weight serving versus NIM, defining post-training data and policy controls, and proving document/agent accuracy on agency corpora before scale. Delivery teams should stage a read-only evaluation, then a bounded post-train pilot with acceptance tests on tables/charts/OCR tasks, and keep Missionforce or other partner stacks optional. Do not treat vendor 5x or Dynamo 8x/85% figures as acceptance metrics. Acceptance criteria below are proposals, not observed results."},"roleDetails":{"sales":{"customerProblem":"Agencies need multimodal agents that read documents and tools under their control without sending sensitive content to closed frontier APIs.","stakeholders":["CIO / Chief AI Officer","Mission/application owners","Security/privacy","Procurement"],"discoveryQuestions":["Which document and workflow classes must stay on-premises?","Is Salesforce Missionforce or another integrator already in scope?","Will evaluation use open weights only or require NIM support?"],"valueHypothesis":"Interpretation: open multimodal agents may cut closed-API spend and improve document automation if post-training works on agency data; this source reports no SLED measured results.","potentialEngagement":"Bounded open-weight evaluation on a sample of agency PDFs/charts plus a written production-path decision (open serving vs NIM).","unsupportedClaims":["Guaranteed 5x throughput in agency environments","That Missionforce government workflows are live with measured outcomes","That Dynamo 8x/85% applies without buyer testing"]},"engineering":{"fit":"High for document-agent and local-agent PoCs on NVIDIA GPUs; uncertain for non-NVIDIA inference estates.","architectureIntegration":"Open weights via HF/vLLM-class stacks or NIM microservice; NeMo post-training; optional Dynamo for multi-GPU serving; Bolt for 24GB local desktops/Sparks.","prerequisites":["GPU capacity and drivers","Document corpus and eval harness","Identity/network isolation for on-prem inference"],"constraints":["Vendor speed and partner metrics unverified here","Bolt E4B not available until Oct 23","Government workflow claim is integrator intent"],"security":"On-prem framing keeps data local; still need RBAC, logging, model-update and prompt-injection controls for agents.","proofOfValue":["Score OCR/table/chart tasks on agency samples","Compare open-weight latency vs any NIM trial","Document data residency decision before production"]},"delivery":{"work":"Stand up evaluation serving, run document-agent tests, optionally post-train with NeMo, then decide production packaging.","dependencies":["GPU ops","Data owners for training samples","Security review","Optional Salesforce/integrator if Missionforce path"],"ownership":"Agency AI platform owner with security co-sign on production.","skillsAdoption":"Need engineers for NeMo/SFT and inference ops; end users need agent workflow training.","governanceCheckpoints":["Eval charter approved","Post-train data inventory cleared","Production license/support path approved"],"acceptanceCriteria":["Proposed: baseline accuracy on named document tasks before/after any post-train","Proposed: written decision on open serving vs NIM"],"risks":["Vendor benchmarks overstate end-to-end agent time","Integrator path bypasses agency architecture review","License cost surprise at production"]}},"retrievedAt":"2026-10-16T02:56:30Z","enrichedAt":"2026-10-16T02:57:15Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Local Bolt models may help agencies keep agent work off cloud for staff without large clusters; skills needed for NeMo post-training and NIM ops.","procurementImplications":"Open weights enable evaluation; production NIM/enterprise support path implies separate software licensing (see pricing resource).","operatingModelImplications":"Shift from closed API agents to adaptable open models that agencies can post-train and host; requires MLOps ownership.","sourceVerification":{"openedUrl":"https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/","referenceExcerpt":"organizations can post-train Super VL on their own data, deploy it on their own infrastructure","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}