# MiaAI-Lab/DeepSeek-v4.1-Flash-DGX-Sparks

> Scripts and patches to serve DeepSeek-V4.1-Flash with SGLang across a 3–4 node NVIDIA DGX Spark cluster, using MXFP4/FP8, speculative decoding and an OpenAI-compatible endpoint.

- Magnitude: 2.0 out of 10 — Steady
- Stars: 199 total · +0 stars on 2026-09-28 (UTC), ≈ 8 by evening
- Star trust: star growth looks organic
- Category: Language models · Language: Python · License: AGPL-3.0 · Created: 2026-09-11 · Last push: 2026-09-25
- GitHub: https://github.com/MiaAI-Lab/DeepSeek-v4.1-Flash-DGX-Sparks · Homepage: https://x.com/MiaAI_lab · Page: https://gitnova.dev/en/r/MiaAI-Lab/DeepSeek-v4.1-Flash-DGX-Sparks

## Useful for

- Deploy an OpenAI-compatible DeepSeek-V4.1-Flash endpoint across several DGX Sparks
- Configure a TP3/TP4 profile with NVMe-resident Engram tables and doctor checks
- Measure throughput and TTFT at 1–4 concurrent streams

## Why it’s here

- 0 stars so far today, about 8 expected by the end of the day.
- The repository is 17 days old and already has 199 stars.
- Recent forks include notable developers: @mrexodia (3,920 followers), @gmh5225 (1,275 followers).

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 34
- Issues and pull requests: 41
- Watchers: 3
- Average over the last week: 8 per day
- Usual pace: 8 per day

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-09-30 … 2026-10-29: None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None

## Magnitude by day, last 14 days

09-15 2.6, 09-16 2.2, 09-17 2.1, 09-18 2.2, 09-19 2.0, 09-20 1.9, 09-21 1.8, 09-22 1.3, 09-23 1.6, 09-24 2.0, 09-25 2.2, 09-26 2.0, 09-27 2.1, 09-28 2.0

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-09-28 00:05 UTC, updated every 30 minutes.