dgx-spark-qwen38
GitHub

dgx-spark-qwen38 / Models / Nemotron 3.5 Bolt

Run Nemotron 3.5 Bolt on a DGX Spark

A 74B that reads images and video, by NVIDIA. With its drafter, on vLLM.

NVIDIA's 74B that reads images and video, in a 24 GB checkpoint, on vLLM with its drafter. On an ASUS Ascent GX10 with 128 GB: 140 tok/s, rewriting code, one request, reasoning off.

Measured

140tok/s

rewriting code, one request, reasoning off

  • 201.2 tok/s in all at 16 streams
  • Reads images and video
  • The lane a 64 GB Spark serves first
Measured on
an ASUS Ascent GX10 with 128 GB, the project's reference box
How
the Bolt lane campaign
Source
docs/nemotron-lane.md

Load it

From the cockpit: open the Lanes view, pick Nemotron 3.5 Bolt and press Load (about 8 minutes to answer). From a terminal, in the repo folder on the box:

./prepare-lane.sh nemotron   # the first time only, builds the lane
./switch-model.sh nemotron   # switches the boot, then prints the commands to run

switch-model.sh never restarts an engine itself: it prints the exact stop and start commands. The cockpit's Load runs them for you.

On a 64 GB DGX Spark

Yes, it is the lane a 64 GB box serves: 110.8 tok/s greedy median and 667,874 tokens of KV, measured at a 64 GB budget on the reference box. What runs on 64 GB

Use it from your tools

The box serves OpenAI- and Anthropic-compatible APIs on port 30001, with an API key. Claude Code, opencode, Open WebUI and the OpenAI and Anthropic SDKs point there instead of a cloud: see Claude Code with a local model and opencode with a local model.

Install it

One command, on stock DGX OS. The same command updates it later.

curl -fsSL https://raw.githubusercontent.com/hasso5703/dgx-spark-qwen38/main/get.sh | bash

Questions about Nemotron 3.5 Bolt on a DGX Spark

Can a DGX Spark run Nemotron 3.5 Bolt?

Yes. NVIDIA's 74B that reads images and video, in a 24 GB checkpoint, on vLLM with its drafter. The installer of dgx-spark-qwen38 sets it up, and the cockpit loads it with one button.

How fast is Nemotron 3.5 Bolt on a DGX Spark?

140 tok/s, rewriting code, one request, reasoning off, measured on an ASUS Ascent GX10 with 128 GB. 201.2 tok/s in all at 16 streams. Reads images and video. The lane a 64 GB Spark serves first.

Does Nemotron 3.5 Bolt run on a 64 GB DGX Spark?

Yes, it is the lane a 64 GB box serves: 110.8 tok/s greedy median and 667,874 tokens of KV, measured at a 64 GB budget on the reference box.

The other lanes