Picom-code/pico-dgx-ds4
Can I trust it?
Worth a closer look before you use it.
Warnings Check
Very new: it was created 11 days ago. Wait a little before installing it.
Stars from real accounts No data
We haven't checked its stars for fakes yet.
Active OK
Last change today.
Licence OK
MIT: you can use it for free, also in a commercial product. Keep the licence notice.
Quick checks from public data. They help, but they are no guarantee.
What it is
A fork of blaesus/dsgx10 that serves DeepSeek V4.1 Flash across four DGX Spark machines for many concurrent users. It adds sparse-attention changes for 1M context, slot admission by context size, speculative drafting, a switch-based collective sync path, and image inputs through…
Who it's for
Engineers running large language models on multi-node DGX Spark or similar GPU clusters who need long context and multi-user serving
How to use it
Ask your AI to set it up
Paste this into Claude, ChatGPT or your coding assistant.
Help me install and set up Picom-code/pico-dgx-ds4 in my project. Follow its README: https://github.com/Picom-code/pico-dgx-ds4. Explain each step and check it is safe before running anything.
Star curve
Alternatives
Similar projects named in its profile. Each link searches the directory.
Get alerts for repos like this
We'll email you when something in Models & inference starts rising. Free, once a week, only what matches.
What arrives: one email on Mondays with the repos that match, never more. See last week's. Privacy
For maintainers
Add this badge to your README. It shows the repo's place on the radar, updates every hour and links back to this page.
[](https://repoalerts.com/repo/Picom-code/pico-dgx-ds4?ref=badge)Similar repos
- basetenlabs/deepseek-ultrafast
An inference engine for DeepSeek-V4.1-Flash on NVIDIA sm_100a and sm_103a GPUs, with 1M-token context and KV cache sharing across servers over RDMA, coordinated through Redis. Includes an OpenAI-compa
- MiaAI-Lab/DeepSeek-v4.1-Flash-1017-2x-DGX-Sparks
Serves the DeepSeek-V4.1-Flash-1017 model split across two NVIDIA DGX Sparks through an OpenAI-compatible API on port 8888. Uses the TensorFold native Zig engine, a 2.9 bpw EXL3 quantization, a 1,048,
- deepseek-ai/DeepInfer
Inference framework for serving DeepSeek models at scale, released as an offline inference build. Provides optimized kernels, attention and MoE parallelism, speculative decoding, asynchronous scheduli
- MiaAI-Lab/GLM-5.3-Flash-DGX-Sparks-1104
Recipe that serves the GLM-5.3-Flash model from an EXL3 checkpoint across two NVIDIA DGX Sparks at tensor parallel 2, using the TensorFold native Zig/CUDA server built in Docker for sm_121. Includes s
- Betelgeuse-AI/Betelgeuse
A llama.cpp fork that runs mixture-of-experts models larger than GPU memory on an RTX 3060 12 GB. Dense weights stay on the GPU, while routed experts are paged from local NVMe into a bounded VRAM slot
- HalfVulpes/Deepinfer-SM80
Ports DeepSeek's DeepInfer kernels and engine optimizations from Blackwell to NVIDIA Ampere SM80/SM86 GPUs for vLLM V1. Ships two vLLM patches plus drop-in Python, Triton and TileLang modules targetin