dgx-spark-qwen38
GitHub

dgx-spark-qwen38 / Models / Qwen3.8-27B

Run Qwen3.8-27B on a DGX Spark

Text and code, a 1M-token window, by Qwen. NVFP4 with a DFlash2 drafter, on SGLang.

Qwen's 27B in NVFP4, served by SGLang with a DFlash2 drafter, with a window of a million tokens. On an ASUS Ascent GX10 with 128 GB: 71.4 tok/s, greedy median, one request.

Measured

71.4tok/s

greedy median, one request

  • 135 to 148 tok/s in all at 8 streams
  • A window of a million tokens
  • GSM8K 94.0% on 200 problems
Measured on
an ASUS Ascent GX10 with 128 GB, the project's reference box
How
./bench.sh, the v1.14.0 campaign
Source
BENCHMARKS.md

Load it

From the cockpit: open the Lanes view, pick Qwen3.8-27B and press Load. From a terminal, in the repo folder on the box:

./switch-model.sh stock   # 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.

Other checkpoints of the same lane load the same way: fp8, uncensored, uncensored-fp8.

On a 64 GB DGX Spark

Yes, without the 1M window: two requests at a time, 59,625 tokens of KV (one prompt of up to 54,855 tokens), 67.2 tok/s greedy median, 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 Qwen3.8-27B on a DGX Spark

Can a DGX Spark run Qwen3.8-27B?

Yes. Qwen's 27B in NVFP4, served by SGLang with a DFlash2 drafter, with a window of a million tokens. The installer of dgx-spark-qwen38 sets it up, and the cockpit loads it with one button.

How fast is Qwen3.8-27B on a DGX Spark?

71.4 tok/s, greedy median, one request, measured on an ASUS Ascent GX10 with 128 GB. 135 to 148 tok/s in all at 8 streams. A window of a million tokens. GSM8K 94.0% on 200 problems.

Does Qwen3.8-27B run on a 64 GB DGX Spark?

Yes, without the 1M window: two requests at a time, 59,625 tokens of KV (one prompt of up to 54,855 tokens), 67.2 tok/s greedy median, measured at a 64 GB budget on the reference box.

The other lanes