> ALPHA-One > ALPHA-One full documentation
Edit page

ALPHA-One full documentation


The ALPHA-One is a Generative 7b LLM AI Agent based on StarPro64 board RISC-V based Single Board Computer with 20TOPS(int 8) NPU capability, and preinstalled 64G eMMC module contains 7b Deepseek/Owen LLM build in docker form.

The ALPHA-One

The ALPHA-One


Software

Releases

Production release build 20250627

Image for 64GB eMMC module

This build contains 7b QWen2 LLM. To access the web UI, open http://alpha-one.local:8000/

Download:

File information:

  • MD5 (GZip file): d62eeb1645c7b496a0815c4590f798c7
  • File Size: 18.10 GB
UserPassword
eswineswin

Image for 32GB eMMC module and above

This is the early test build that contains both 7b Deepseek and QWen2 LLM build, OS based on ROCKOS. For larger than 32GB eMMC, please use gpart to extend the file partition.

Download:

File information:

  • MD5 (GZip file): 6669a436104e97e2ad90679e06f62d78
  • File Size: 18.21 GB
UserPassword
eswineswin

Usage

To explore LLM, use terminal program and execute below command lines:

DeepSeek 7b LLM command:

$ sudo /opt/eswin/sample-code/npu_sample/qwen_sample/bin/es_qwen2 /opt/eswin/sample-code/npu_sample/qwen_sample/src/deepseek_r1_distill_qwen_7b/config.json

Qwen2 7bLLM command:

$ sudo /opt/eswin/sample-code/npu_sample/qwen_sample/bin/es_qwen2 /opt/eswin/sample-code/npu_sample/qwen_sample/src/qwen2_7b/config.json

Qwen2 0.5bLLM command:

$ sudo /opt/eswin/sample-code/npu_sample/qwen_sample/bin/es_qwen2 /opt/eswin/sample-code/npu_sample/qwen_sample/src/qwen2_0_5b/config.json

SDKs

The SDK documents can be found under https://files.pine64.org/SDK/StarPro64/docs.7z (some are in Chinese).

File information:

  • MD5 (7-Zip file): 8ee9b660c336d9979191d212e9826226
  • File Size: 34 MB

SDK Release:

LLM and tools release:


Specifications

Main Board and SoC Specification

EIC7700X block diagram

EIC7700X block diagram

Features

System Memory

  • 32GB 64bits LPDDR5@6400MHz RAM Memory.
  • 64GB eMMC preinstalled with 7b Deepseek/Qwen LLM

Case

  • CNC from single Aluminum block
  • Passive cooling, built-in heat pipe that directly contact with SoC

Network

  • Dual 10/100/1000 Mbps Ethernet
  • 2.4GHz/5GHz MIMO WiFi 802.11 b/g/n/ac/ax with Bluetooth 5.3

Expansion Ports

  • 2× USB3.0 Dedicated Host port
  • 2× USB2.0 Shared Host port
  • Digital Video output up to 4K@60Hz
  • 3.5mm audio Jack with mic in

Certifications

  • Not yet available

Potential Improvement

  • Current 7bLLM throughput is around 3.5 tokens/second using docker form, potential improve up to 10 tokens/second when operates LLM in native mode.
  • Potential increase LLM capability from 7b to 15b with system software optimization
  • If ALPHA-One proven its potential, consider release ALPHA-Two which can host up to 30b LLM