NVIDIA Nemotron 3.5 brings a 1M-context agent model and 74B local variant

What changed · following since Oct 16 Oct 20 — NVIDIA positioned Nemotron 3.5 for private enterprise data centers and local agents on RTX Spark PCs and DGX Spark workstations.

The release now spans a capability-focused multimodal tier and a smaller path to local execution, rather than treating agent deployment as a single model-sizing problem. Super VL targets long-context multimodal work, while Bolt trades scale for a score close to Nemotron 3 Ultra.

• Nemotron 3.5 Super VL is a 120-billion-parameter mixture-of-experts model with 12 billion active parameters, a 1-million-token context window, vision and text input, and text output.
• Nemotron 3.5 Bolt 74B-A9B scored 21 on the Artificial Analysis Intelligence Index, compared with 23 for the 550B-parameter Nemotron 3 Ultra; Bolt has about one-seventh the total parameters and one-sixth the active parameters.
• Artificial Analysis scored Nemotron 3.5 Super VL at 26.5, ahead of Thinking Machines’ Inkling Small at 25.7 and NVIDIA’s Nemotron 3 Ultra at 23; it also reported twice the score of Nemotron 3 Super while using about 60% fewer output tokens per task.
• vLLM announced day-one support for Nemotron 3.5 Super VL on October 15, including OpenAI-compatible serving, multimodal inputs, and speculative decoding for long-running agent workflows.
• Ollama v0.40.5, released on October 14 at 19:45, lists NVIDIA Nemotron 3.5 Super as a supported vision-language model for coding, reasoning, tool use, and document understanding.
• NVIDIA positioned the Nemotron 3.5 models for enterprise data centers and local agents on RTX Spark PCs and DGX Spark workstations, with systems entering the market through partners including Microsoft and Dell.

Bottom line If you build multimodal or always-on agents, benchmark Super VL and Bolt against your latency, memory, and tool-use requirements before choosing a serving path.