AMEVA-Vulkan-Runtime

SoC-Aware Adaptive Abstraction Runtime Utilizing Device Resources for Android Termux

1-Line Quick Installation

Install the official package directly into your runtime:

pip install ameva-vulkan-runtime
# or: npm install ameva-vulkan-runtime

The Engineering Challenge

Running multi-modal AI on mobile Android requires adaptive utilization of device resources across diverse SoC architectures (Qualcomm Adreno and Samsung Exynos ARM Mali).

The Architectural Breakthrough

Provides a C++20 Hardware Abstraction Layer (HAL) with Zero-Guesswork SoC auto-detection, utilizing hardware paths on Qualcomm Adreno and stable ARM NEON 4-Thread FP16 CPU execution on Exynos Mali.

Key Capabilities & Built-in Hardening

Runtime SoC Auto-Detection

Scans /proc/cpuinfo and device nodes on startup to route execution without guesswork.

Adaptive Resource Routing

Routes to hardware paths on Adreno and deterministic ARM NEON 4-Thread FP16 CPU execution on Exynos Mali.

Unified Resource Core

Powers Whisper STT, LLaVA Vision, LLaMA/BitNet LLM, and Stable Diffusion from a single shared core.

Supported Compute Kernels & Operations

Subsystem Category	Operations & Kernels	Status
Target SoCs	Qualcomm Snapdragon (Adreno), Samsung Exynos (Cortex CPU NEON & Mali)	Production
Operation Modes	Hardware Execution (Adreno), ARM NEON 4-Thread FP16 CPU Mode (Exynos)	Production
Roadmap	Phase 1 CLI Deterministic Routing, Phase 2 Android Foreground Service JNI Wrapper	Production

Canonical Usage Example

import ameva_vulkan_runtime as avr

# 1. Probe & Validate Hardware (V0-V11)
doctor = avr.Doctor()
report = doctor.run_self_test()
print(f"GPU Backend Status: {report.overall_success}, Device: {report.device_name}")

# 2. Acquire High-Performance Vulkan Context for STT / LLM / Diffusion
ctx = avr.create_context(device="auto", memory_limit_mb=1024)
print(f"Context initialized via: {ctx.loader_path} (API {ctx.driver_version})")

import { Doctor, createContext } from "ameva-vulkan-runtime";

const doctor = new Doctor();
const report = await doctor.runSelfTest();
console.log(`GPU Status: ${report.overallSuccess}, GPU: ${report.deviceName}`);

const ctx = await createContext({ device: "auto" });
console.log(`Vulkan Context Ready on ${ctx.deviceName}`);