Skip to content
View BHCC2025's full-sized avatar

Block or report BHCC2025

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
BHCC2025/README.md

BHCC2025 · NVIDIA DGX Spark LLM Setup — Start Here

Run open LLMs such as Qwen and Gemma locally on 1, 2 or 3 NVIDIA DGX Sparks (GB10) with vLLM: tested recipes, one setup script, and multi-Spark tensor parallelism over direct QSFP cables.

➜ Start here: the NVIDIA DGX Spark LLM setup guide

What to do first, how to cable two or three Sparks, and which recipe to pick, with measured speeds.

Repositories

Popular repositories Loading

  1. sparkDash sparkDash Public

    Forked from MiaAI-Lab/sparkDash

    sparkDash ⚡ — Multi-DGX Spark Monitoring Dashboard

    JavaScript

  2. DeepSeek-v4.1-Flash-DGX-Sparks DeepSeek-v4.1-Flash-DGX-Sparks Public

    Forked from MiaAI-Lab/DeepSeek-v4.1-Flash-DGX-Sparks

    DeepSeek-V4.1-Flash on 3-4x NVIDIA DGX Sparks

    Python

  3. dgx-spark-recipe-kit dgx-spark-recipe-kit Public

    Shared setup kit for DGX Spark recipe repos: node/SSH/dependency checks, cabling detection, fabric IPs, NCCL network test, model staging

    Shell

  4. Qwen3.8-Flash-Next-DGX-Spark-TP1-TP3 Qwen3.8-Flash-Next-DGX-Spark-TP1-TP3 Public archive

    Archived: vLLM recipe for Qwen3.8 Flash Next on 1-3 DGX Sparks. We now use TensorFold; see NVIDIA-DGX-Spark-LLM-Setup.

    Python

  5. Gemma-4-31B-IT-DGX-Spark-TP1-TP2 Gemma-4-31B-IT-DGX-Spark-TP1-TP2 Public

    Gemma-4-31B-IT (NVFP4) on 1 or 2 NVIDIA DGX Sparks with vLLM: TP1/TP2, draft-model speculative decoding, benched (40.6 tok/s code on 2 Sparks)

    Shell

  6. BHCC2025 BHCC2025 Public

    NVIDIA DGX Spark LLM setup — start here: tested vLLM recipes for Qwen and Gemma on 1, 2 or 3 DGX Sparks (GB10), one setup script