OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Infrastructure & Serving

Sndr Core Engine

Sndr Core Engine is a Python-based, Apache-2.0 runtime patch overlay for vLLM that optimizes local inference and serving of supported open LLMs, with deeper tuning for Qwen3.6 and Gemma 4. It includes a CLI and a browser-based Control Center for launching and managing the engine.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Sndr Core Engine — OpenViST project profile in AI Infrastructure & Serving.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want to self-host supported open LLMs with vLLM on consumer NVIDIA hardware and can provide the documented Linux, CUDA and Docker stack.

Who it is for

  • Homelab operators hosting open models on consumer NVIDIA GPUs.
  • On-premise teams running inference on their own Linux hardware.
  • Agent builders who need a local model-serving endpoint and tool calling.
  • Researchers and performance engineers examining runtime patches and benchmarks.

Documented uses

  • Serve supported Qwen3.6 and Gemma 4 models locally through vLLM.
  • Launch a local inference engine and manage it through the Control Center.
  • Use documented runtime optimizations such as quantized KV cache and speculative decoding.
  • Inspect patch and benchmark documentation when evaluating inference behavior.
Before you choose

Requirements & limits

The repository specifies an Apache-2.0 license.

  • A local engine requires Linux, CUDA and Docker; Mac and Windows are documented as remote-client options for a Linux rig.
  • The project lists consumer NVIDIA cards including RTX 3090, 4090 and 5090, plus A5000 and A6000; consult its hardware guidance for configuration-specific constraints.
  • The deeper optimizations are tuned for Qwen3.6 and Gemma 4, while the README says vLLM serves models supported by the engine.
  • The README's speed and context figures are project claims, not independent performance guarantees.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Sndr Core and launch the Control Center

Install the CLI and start the engine with the Control Center on a documented Linux host.

Before you start

  • Linux
  • CUDA
  • Docker
  • A supported NVIDIA GPU configuration

Steps

  1. Run the install command shown in the README: the command below.

    Documented command
    curl -sSL https://raw.githubusercontent.com/Sandermage/sndr_core_engine/main/install.sh | bash

  2. Run sndr up to select a GPU preset, download model weights if needed, and launch the engine and Control Center.

  3. Open the Control Center at http://127.0.0.1:8765.

Expected outcome

The engine and local browser-based Control Center are started.

Guide limitations

  • The local engine requires Linux, CUDA and Docker.
  • GPU-specific constraints and presets vary; consult the project's hardware documentation and FAQ.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for Model inference and serving

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
Sndr Core EngineServe supported Qwen3.6 and Gemma 4 models locally through vLLM.Linux, Self-hostedThe repository specifies an Apache-2.0 license.
MorphingdbImport a libtorch model and create predictions from values in a PostgreSQL table.Docker, Linux, Self-hosted, ServerThe README documents a Docker image and a build-from-repository option, followed by connecting to PostgreSQL and enabling the pgdl extension.
Gateway — pyadavSend chat-completion requests to a provider through a common server endpoint.ServerThe repository is public and identifies the project as Apache-2.0 licensed. README installation instructions include binaries, platform package managers and Docker.
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
Sandermage/sndr_core_engine · Official homepage title · retained in its source language
How we review sourcesSuggest a correction