RepoAlerts

basetenlabs/deepseek-ultrafast

App

Can I trust it?

Be careful with this one.

  • Stars from real accounts No data

    We haven't checked its stars for fakes yet.

  • Active OK

    Last change 2 days ago.

  • Licence Careful

    No licence: legally you can't reuse the code, unless the README says otherwise.

Quick checks from public data. They help, but they are no guarantee.

What it is

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-compatible server.

Who it's for

Engineers deploying large-context DeepSeek model inference on multi-GPU NVIDIA B200, B300 or GB300 servers

How to use it

Install

DSUF_SM=sm_103a cargo build --release -p deepseek-ultrafast-engine --bin run -p deepseek-ultrafast-server --bin serve

From its README. Check it there before running it.

Ready-made installers, when the project publishes them, are on its latest release.

Downloads (if any)

Ask your AI to set it up

Paste this into Claude, ChatGPT or your coding assistant.

Help me install and set up basetenlabs/deepseek-ultrafast in my project. Follow its README: https://github.com/basetenlabs/deepseek-ultrafast. Explain each step and check it is safe before running anything.

Star curve

438From 0 to 438 stars between 27 Oct 2026 and 16 Nov 2026.

On the radar

14 Nov 2026: #7515 Nov 2026: #28
Best place #28, on 15 Nov 2026. In the top 100 on 2 of the last 30 days.

Its place among the radar's top 100, day by day, with #1 at the top. Gaps are days it was out of the top 100.

Records

  • Best place on RepoAlerts: #28, on 15 Nov 2026
  • In the top 100 for 2 days since 14 Nov 2026

Alternatives

Similar projects named in its profile. Each link searches the directory.

On Hacker News

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.

[![basetenlabs/deepseek-ultrafast on RepoAlerts](https://repoalerts.com/badge/basetenlabs/deepseek-ultrafast.svg)](https://repoalerts.com/repo/basetenlabs/deepseek-ultrafast?ref=badge)
Share on X

  • 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

    LibraryModels & inferencePython · 26,098 stars · +377 this week · 31 days old · HN 2351 pts

  • 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,

    AppModels & inferencePython · 483 stars · +43 this week · 29 days old

  • Hyperstitioncc/Hyperkernel

    Rust implementations of megakernels for NVIDIA B300 GPUs, with host servers for DeepSeek-V4.1-Flash LLM serving, openpilot driving-model inference, Evo 2 7B DNA generation and Matrix-Game 3.0 world-mo

    AppModels & inferenceRust · 199 stars · +16 this week · 11 days old

  • 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

    LibraryModels & inferencePython · 200 stars · 27 days old

  • Magnvs7/Promethevs

    A local inference engine that runs the full DeepSeek-V4.1-Flash-1017 model from its original 510 GB weights on a PC with 64 GB RAM and a 12 GB NVIDIA GPU. It uses a 2-bit expert copy and recomputes un

    AppModels & inference190 stars · +34 this week · 26 days old · HN 5 pts

  • deepseek-ai/deepseek-recipe

    A collection of Rust libraries and Python bindings that convert Messages, Chat Completions and Responses API requests into a common Conversation format, render it into DeepSeek V4 and V4.1 prompts, en

    LibraryModels & inferenceRust · 2,878 stars · +68 this week · 67 days old