# espressif/esp-dl

Espressif deep-learning library for AIoT applications

Repository: https://github.com/espressif/esp-dl
Canonical: https://ross.abutalabs.com/products/esp-dl
Language: Assembly
License: MIT
License Family: permissive
Last push: 2026-08-31T07:22:46+00:00

## Health v2 (maintenance only)
Score: 72/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 100, release rhythm 21, longevity 100
- inputs: {"age_days": 2847, "days_push": 2, "days_rel": 314, "gap_med": 306, "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 1125, forks 222 (observed 2026-09-01T02:14:01.246428+00:00)

## What it is
ESP-DL is Espressif's lightweight neural network inference framework for ESP-series chips, with a custom .espdl model format, quantization tooling (ESP-PPQ), and optimized operators for on-device AI. It includes a static memory planner, dual-core scheduling, and prebuilt models such as OCR for running inference directly on microcontrollers.

## Use cases
- run neural network inference on esp32
- deploy quantized models on microcontrollers
- on-device ocr for embedded systems
- build aiot applications with esp chips
- convert onnx pytorch tensorflow models for esp32
- image recognition on edge devices

## When to choose
- you are developing AI applications on Espressif ESP32-series SoCs
- you need efficient, memory-optimized inference on resource-constrained hardware
- you want to quantize standard models (ONNX, PyTorch, TensorFlow) for embedded deployment

## When to avoid
- you need training or fine-tuning rather than inference
- your target hardware is not an Espressif chip
- you need a full-featured deep learning framework like PyTorch or TensorFlow

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, llm-inference, image-processing, ocr, computer-vision
- domain: machine-learning, deep-learning, iot, embedded-systems, artificial-intelligence
- platform: embedded, iot, cpp, c
- tags: esp32, neural-network-inference, model-quantization, on-device-ai, flatbuffers, aiot, microcontroller, edge-ai

## Member repositories
- espressif/esp-dl (main) score 72

## Provenance
- Observed fields: from GitHub, fetched 2026-09-01T02:14:01.246428+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 2026-08-30T06:39:38.505153+00:00, confidence not recorded.
  - readme: https://github.com/espressif/esp-dl (fetched 2026-09-01T02:14:01.246428+00:00, sha bf3a86f861fe)
- Data as of 2026-08-30T08:39:29.467469+00:00.