# Xilinx/Vitis-AI

Vitis AI is Xilinx’s development stack for AI inference on Xilinx hardware platforms, including both edge devices and Alveo cards.

Repository: https://github.com/Xilinx/Vitis-AI
Canonical: https://ross.abutalabs.com/products/vitis-ai
Homepage: https://www.xilinx.com/ai
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-02-24T04:12:06+00:00

## Health v2 (maintenance only)
Score: 71/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 69, release rhythm 56, longevity 100
- inputs: {"age_days": 2513, "days_push": 190, "days_rel": 212, "gap_med": 40, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1800, forks 675 (observed 2026-08-28T04:05:37.946526+00:00)

## What it is
AMD Vitis AI is an integrated development environment and stack for accelerating AI inference on AMD/Xilinx adaptable platforms, including adaptive SoCs (edge devices) and Alveo data center accelerator cards. It provides optimized IP, tools, libraries, pre-trained models, example designs, and tutorials covering the full inference deployment workflow.

## Use cases
- deploy deep learning models on FPGA hardware
- accelerate AI inference on Xilinx edge devices
- run quantized neural networks on Alveo cards
- optimize model inference for adaptive SoCs
- compile ML models for FPGA acceleration
- benchmark inference performance on AMD adaptive hardware

## When to choose
- you are deploying AI inference on Xilinx/AMD FPGAs, adaptive SoCs, or Alveo accelerator cards
- you need quantization and compilation tools to map models onto programmable hardware
- you want pre-optimized models and example designs for edge or data center acceleration

## When to avoid
- your target hardware is AMD Instinct/Radeon GPUs - use ROCm instead
- you need general-purpose GPU training rather than inference acceleration on adaptable hardware
- you do not own any Xilinx/AMD adaptive hardware

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, llm-inference, sdk, developer-tools
- domain: machine-learning, deep-learning, embedded-systems, developer-tools
- platform: python, cpp, embedded
- tags: fpga, xilinx, amd, ai-inference, edge-ai, alveo, quantization, model-zoo, hardware-acceleration, linux, gpu

## Member repositories
- Xilinx/Vitis-AI (main) score 71

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:37.946526+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:22:14.400483+00:00, confidence not recorded.
  - readme: https://github.com/Xilinx/Vitis-AI (fetched 2026-08-28T04:05:37.946526+00:00, sha ebb35a8d5bc1)
  - homepage: https://www.xilinx.com/ai (fetched 2026-08-29T11:01:10.263720+00:00, sha d70f07723132)
  - site_page: https://www.amd.com/en/search/documentation/hub.html (fetched 2026-08-29T11:01:10.266671+00:00, sha 6e948b307a13)
  - site_page: https://rocm.docs.amd.com/en/latest (fetched 2026-08-29T11:01:10.268804+00:00, sha ad05089c40ff)
  - site_page: https://www.amd.com/en/partner/about-partners/partner-ecosystems-solutions.html (fetched 2026-08-29T11:01:10.274708+00:00, sha 96bbf5403ca8)
  - site_page: https://www.amd.com/en/partner/about-partners/authorized-distributors.html (fetched 2026-08-29T11:01:10.276583+00:00, sha 6b77c73858bc)
  - site_page: https://www.amd.com/en/partner/browse-by-resource/faqs.html (fetched 2026-08-29T11:01:10.272400+00:00, sha 48def8afba03)
  - site_page: https://www.amd.com/en/corporate.html (fetched 2026-08-29T11:01:10.278284+00:00, sha 0206f5583b9e)
- Data as of 2026-08-30T08:39:29.467469+00:00.