# lucidrains/perceiver-pytorch

Implementation of Perceiver, General Perception with Iterative Attention, in Pytorch

Repository: https://github.com/lucidrains/perceiver-pytorch
Canonical: https://ross.abutalabs.com/products/perceiver-pytorch
Language: Python
License: MIT
License Family: permissive
Topics: artificial-intelligence, deep-learning, attention-mechanism
Last push: 2026-06-08T02:26:07+00:00

## Health v2 (maintenance only)
Score: 62/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 86, release rhythm 8, longevity 100
- inputs: {"age_days": 2007, "days_push": 87, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1217, forks 140 (observed 2026-08-28T04:04:01.412900+00:00)

## What it is
A PyTorch implementation of the Perceiver architecture (General Perception with Iterative Attention) and its follow-up Perceiver IO. It provides ready-to-use model classes for processing multimodal inputs like images and video via cross-attention to a latent array.

## Use cases
- implement perceiver architecture in pytorch
- train a model on images with iterative attention
- use perceiver io for flexible output sequences
- experiment with cross-attention on multimodal data
- reproduce deepmind perceiver paper results

## When to choose
- you want a pip-installable PyTorch implementation of Perceiver or Perceiver IO
- you need a research baseline for latent-attention architectures
- you want to experiment with attention over arbitrary input modalities

## When to avoid
- you need a production-trained model with pretrained weights
- you require a full training pipeline or dataset tooling
- you need architectures beyond Perceiver, like standard transformers

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning
- domain: deep-learning, machine-learning, artificial-intelligence
- platform: python
- tags: perceiver, attention-mechanism, transformers, pytorch, research-implementation

## Member repositories
- lucidrains/perceiver-pytorch (main) score 62

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:01.412900+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 2026-08-30T06:17:08.234669+00:00, confidence not recorded.
  - readme: https://github.com/lucidrains/perceiver-pytorch (fetched 2026-08-28T04:04:01.412900+00:00, sha eff58d7dce17)
- Data as of 2026-08-30T08:39:29.467469+00:00.