# MiaAI-Lab/DeepSeek-v4.1-Flash-1017-2x-DGX-Sparks

> Scripts to serve the quantized DeepSeek-V4.1-Flash-1017 (EXL3, 2.9 bpw) across two NVIDIA DGX Sparks via the TensorFold engine, exposing an OpenAI-compatible API with up to a 1M-token window and DSpark speculative decoding.

- Magnitude: 4.9 out of 10 — Early signal
- Stars: 118 total · +9 stars measured 2026-10-20, 10:12–12:25 UTC, ≈ 35 by evening
- Star trust: star growth looks organic
- Category: Language models · Language: Shell · License: Apache-2.0 · Created: 2026-10-18 · Last push: 2026-10-20
- GitHub: https://github.com/MiaAI-Lab/DeepSeek-v4.1-Flash-1017-2x-DGX-Sparks · Homepage: https://mia-ai.net · Page: https://gitnova.dev/en/r/MiaAI-Lab/DeepSeek-v4.1-Flash-1017-2x-DGX-Sparks

## Useful for

- Bring up an OpenAI-compatible DeepSeek endpoint on two DGX Sparks with a single ./start.sh
- Run prefill and decode benchmarks via tools/bench.py, including a 479K-token needle-in-a-haystack test
- Evaluate the 2.9-bit EXL3 quantization quality on contexts up to 512K tokens

## Why it’s here

- Star-counter measurements on 2026-10-20 (UTC), 10:12–12:25: 109 → 118 stars (+9). This is the change over that interval.
- Estimated end-of-day forecast: about +35 stars, using observed gains and the previous day.
- The repository is 2 days old and already has 118 stars. With less than two weeks of history, there's no usual pace to compare the spike against yet.
- Top new repositories this week: #180.
- About 20 forks a day — people are taking the code.

## 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: 19
- Issues and pull requests: 18
- Watchers: 2
- Average over the last week: 44 per day
- Usual pace: too little history (under two weeks)
- Stars in the last hour (measured): 3

## Stars per day, last 3 days (oldest → newest, today is partial)

2026-10-18 … 2026-10-20: 0, 97, 9

## Spotted in now

- Top new repositories this week: #180

## Similar by description

1. **mw00/project-maya** — 3.2 · Peaking · Language models · C++ · +91 stars measured 2026-10-20, 00:27–12:27 UTC, ≈ 154 by evening
   Local engine, server and dashboard to run the 321B MoE model GLM-5.3-Flash on one or two NVIDIA GPUs, offloading experts to RAM and NVMe.
   Full card: https://gitnova.dev/en/r/mw00/project-maya.md

---

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 2026-10-20 12:38 UTC, updated every 30 minutes.