{"schema":"postcutoff/event@1","as_of":"2026-10-15T23:34:00+02:00","url":"https://postcutoff.com/e/2026-10-15-nvidia-nemotron-3-5-super-vl-bolt/","md":"https://postcutoff.com/e/2026-10-15-nvidia-nemotron-3-5-super-vl-bolt/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-10-15-nvidia-nemotron-3-5-super-vl-bolt","date":"2026-10-15","date_precision":"day","short_title":"NVIDIA releases open Nemotron 3.5 Super VL and Bolt models","deck":null,"takeaway":"On Oct 15, 2026 NVIDIA released Nemotron 3.5 Super VL, an open (OpenMDW-1.1) 120B-total / 12B-active hybrid Mamba-Transformer MoE vision-language model with up to 1M tokens of context, for enterprise agents that read documents, charts and web pages.","category":"open-source","category_label":"Open source","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":[{"n":1,"title":"NVIDIA blog (Kari Briski): Open from data center to desktop, Nemotron Super VL and Bolt","url":"https://blogs.nvidia.com/blog/nemotron-3-5-super-vl/","type":"official","group":"primary","domain":"blogs.nvidia.com"},{"n":2,"title":"NVIDIA Technical Blog: Nemotron 3.5 Super VL powers specialized agents","url":"https://developer.nvidia.com/blog/nvidia-nemotron-3-5-super-vl-powers-specialized-agents-with-multimodal-reasoning/","type":"official","group":"primary","domain":"developer.nvidia.com"},{"n":3,"title":"Hugging Face: NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4","url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4","type":"code","group":"primary","domain":"huggingface.co"},{"n":4,"title":"VentureBeat: Nemotron 3.5 Super VL with post-training beats Claude Opus 5, GPT-6.1 Sol on specialized task benchmarks","url":"https://venturebeat.com/technology/nvidias-new-open-nemotron-3-5-super-vl-with-post-training-beats-claude-opus-5-gpt-6-1-sol-on-specialized-task-benchmarks","type":"press","group":"press","domain":"venturebeat.com"}],"official":3,"filed":"2026-10-15","updated":"2026-10-15","orgs":["NVIDIA"],"title":"NVIDIA releases open Nemotron 3.5 Super VL (120B-A12B) for document agents, plus Nemotron 3.5 Bolt models built for 24 GB and 4 GB devices","summary":"On Oct 15, 2026 NVIDIA released Nemotron 3.5 Super VL, an open (OpenMDW-1.1) 120B-total / 12B-active hybrid Mamba-Transformer MoE vision-language model with up to 1M tokens of context, for enterprise agents that read documents, charts and web pages. It also released Nemotron 3.5 Bolt 74B-A9B, natively 2-bit to fit in 24 GB for RTX PCs, with a 4 GB Bolt E4B due Oct 23. NVIDIA's claim that post-trained versions beat Claude Opus 5 and GPT-6.1 Sol applies only to narrow custom tasks.","key_facts":["Super VL: 120B total / 12B active; Mamba-2–Transformer hybrid MoE with C-RADIO vision encoder and multi-token prediction; context up to 1M; OpenMDW-1.1 licence (Hugging Face card)","Training: ~26T tokens of pretraining and continued pretraining, ~3.8M post-training samples, RL in ~100 open environments","VentureBeat: on NVIDIA's chart, post-training lifted medical-claims accuracy from ~28% to ~72% (Claude Opus 5 ~38%) and security-attack resilience from ~35% to ~76% (GPT-6.1 Sol ~66%); task definitions and sample sizes not published","Artificial Analysis (via VentureBeat): ~625 output tok/s at 100K input on 4× B200 at an index score of ~26, below GLM-5.3-Flash (42) and Qwen3.8-Flash-Next (40)","Price on OpenRouter at launch: $0.15 input / $0.50 output per 1M tokens, plus a free tier (nvidia/nemotron-3.5-super-vl)","Bolt 74B-A9B 24GB: native 2-bit, for RTX Spark laptops (on sale Oct 16), GeForce RTX, DGX Spark; NVIDIA says #1 open model in class on PinchBench, #2 on BrowseComp (no scores given)","Bolt E4B 4GB for voice assistants and device agents due Oct 23","Partners customizing: Scale AI, AI21, Mercor, Ramp, Lila Sciences and others; integrating: Atlassian, CrowdStrike, Perplexity, Salesforce, Synopsys"],"key_numbers":[],"tags":["llm","open-weights","multimodal","agents","mixture-of-experts","mamba","on-device","nvidia"],"science":null,"body_md":"## What happened\n\nNVIDIA expanded its open Nemotron 3.5 family on Oct 15, 2026. The main model, **Super VL**, is a 120B / 12B-active vision-language\nmodel for agents that must read tables, charts, PDFs and web pages and call tools. NVIDIA released it with training recipes and data\nfor customization, and showed post-trained variants beating closed frontier models on five narrow tasks. VentureBeat noted that these\nare customized models on specific tasks, not evidence that the base model beats Claude Opus 5 or GPT-6.1 Sol in general.\n\nThe **Bolt** models target local PCs: a 74B-A9B model compressed natively to 2 bits so it fits in 24 GB, timed for the RTX Spark\nlaptops that go on sale Oct 16, and a 4 GB model due Oct 23.\n\n## Why it matters\n\nNVIDIA is now one of the main US sources of open weights, alongside Meta's Muse Spark Open released the same day. Its pitch is cheap,\nfast, customizable agents, rather than topping general leaderboards.","disputed":[],"related":[{"id":"2026-10-07-microsoft-windows-rtx-spark-event-mxc","url":"https://postcutoff.com/e/2026-10-07-microsoft-windows-rtx-spark-event-mxc/","date":"2026-10-07","date_precision":"day","short_title":"Microsoft and Nvidia launch RTX Spark Windows PCs for local AI","deck":"Microsoft Execution Containers for sandboxing AI agents reach general availability","takeaway":"Agents with computer access have caused real incidents (sandbox escapes, data leaks).","category":"hardware-compute","category_label":"Chips & compute","importance":3,"confidence":"medium","status":{"key":"partly","labels":["Partly confirmed"]},"sources":6,"official":2,"filed":"2026-10-08","updated":"2026-10-08","orgs":["Microsoft","Nvidia"]},{"id":"2026-08-11-nvidia-nemotron-3-5-lightning","url":"https://postcutoff.com/e/2026-08-11-nvidia-nemotron-3-5-lightning/","date":"2026-08-11","date_precision":"day","short_title":"NVIDIA releases open Nemotron 3.5 Lightning and NeMo Switchyard model router","deck":null,"takeaway":"NVIDIA is now a significant American open-weight model developer; cheap local MoE workers plus routing directly target the cost of long-running agents, which dominate 2026 inference demand.","category":"open-source","category_label":"Open source","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":5,"official":3,"filed":"2026-09-29","updated":"2026-09-29","orgs":["NVIDIA"]}],"people":[],"posts":[],"videos":[{"id":"nvidia-developer-introducing-nemotron-3-5-super-vl","title":"Introducing NVIDIA Nemotron 3.5 Super VL","url":"https://postcutoff.com/v/nvidia-developer-introducing-nemotron-3-5-super-vl/"}],"models":[{"id":"nemotron-3-5-super-vl","name":"NVIDIA Nemotron 3.5 Super VL (120B-A12B)","url":"https://postcutoff.com/m/nemotron-3-5-super-vl/"}],"changes":[{"date":"2026-10-15","type":"filed","text":"Created. </content> </invoke>"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-10-15","run":null,"sources_read":null,"updated":"2026-10-15","human_review":null,"version":{"date":"2026-10-15"}},"gaps":[{"model_id":"gpt-6-astra","name":"GPT-6 Astra","cutoff":"2026-04","days_after":168,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":107,"in_training_data":false},{"model_id":"gemini-4-argon","name":"Gemini 4 Argon","cutoff":"2026-03","days_after":198,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":137,"in_training_data":false}],"short_url":null}