ovkit logo

ovkit

OpenVINO inference in 3 lines. One Model class, clean Results, 30+ ready models — with AUTO/NPU/GPU devices, async throughput, and INT8.

CI Docs Python License PyPI Models

Docs · 한국어 문서 · Model catalog · Examples

from ovkit import Model

r = Model("detect")("image.jpg")[0]   # download -> convert -> cache -> run
r.save("out.jpg")                     # boxes drawn; r.boxes.xyxy / .conf / .cls

Or without writing Python at all:

ovkit run detect image.jpg            # prints results, saves image_out.jpg

Install

pip install ovkit

Extras: ovkit[quant] (INT8/NNCF) · ovkit[genai] (LLM/STT) · ovkit[all]. Python 3.10+. For development, install from source:

git clone https://github.com/leeyunjai82/ovkit.git && cd ovkit
python -m venv .venv && source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install -e ".[dev]"

Supported tasks

Every task below runs end-to-end through the same 3 lines — swap the alias:

Task Alias Output Example
Object detection detect, face_detection, person_detection, vehicle_detection, text_detection, license_plate r.boxes detect.py
Classification classify, person_attributes, vehicle_attributes r.probs / text classify.py
Segmentation segment, instance_segmentation r.masks segment.py
Pose / landmarks pose, face_landmarks r.keypoints pose.py
Face analysis age_gender, emotion, head_pose, face_reid r.text (e.g. "age 31 · male 98%") face_analysis.py
OCR text_recognition r.text ocr.py
Super-resolution super_resolution upscaled image via r.plot() super_resolution.py
LLM / STT (GenAI) llm, stt generated text llm.py / stt.py
NLP / audio / time series qa, translation, noise_suppression, time_series tensors via model.infer() denoise_audio.py

The registry exposes one well-tested model per capability (35 total); the HF mirror hosts the full Apache-2.0 OMZ set (other tiers, int8, sparse variants) — surfacing a variant is a one-line edit (catalog). ovkit list shows everything with descriptions.

Devices

Device How Notes
AUTO (default) Model("detect") OpenVINO picks the best device
CPU Model("detect", device="CPU") works everywhere
GPU device="GPU" Intel iGPU / Arc
NPU device="NPU" Intel® Core™ Ultra AI accelerator

Single images run synchronously; stream=True uses an AsyncInferQueue for video/webcam throughput. INT8: model.quantize(calib_images) (NNCF).

Benchmarks

python scripts/benchmark.py        # prints a paste-ready CPU/GPU/NPU table
model CPU GPU NPU
rtdetr_r50 429.4 ms (2 FPS) 36.9 ms (27 FPS) —*
face_detection_0205 11.7 ms (85 FPS) 4.5 ms (224 FPS) —*
person_detection_0202 13.4 ms (75 FPS) 4.7 ms (211 FPS) 6.6 ms (151 FPS)
resnet50_binary_0001 7.9 ms (126 FPS) 5.1 ms (195 FPS) —*
road_segmentation_adas_0001 28.6 ms (35 FPS) 13.0 ms (77 FPS) 23.8 ms (42 FPS)
human_pose_estimation_0007 82.2 ms (12 FPS) 14.0 ms (71 FPS) 22.7 ms (44 FPS)
age_gender_recognition_retail_0013 0.5 ms (1867 FPS) 0.5 ms (2098 FPS) 0.6 ms (1569 FPS)

Measured on an Intel® Core™ Ultra (Lunar Lake) laptop — CPU / integrated GPU / NPU, median of 30 runs, 1280x720 input, OpenVINO 2026.3. — = model not supported by the NPU compiler (dynamic shapes or unsupported ops).

Usage

Python 
from ovkit import Model

model = Model("face_detection")              # alias, name, .xml, or .onnx
results = model("photo.jpg", conf=0.25)      # image / ndarray / folder / video
for r in model.predict(0, stream=True):      # webcam (lazy generator)
    annotated = r.plot()

print(Model("age_gender")("face.jpg")[0].text)   # "age 31 · male 98%"

Inputs are auto-detected: image path / ndarray / folder / video / camera index → vision pipeline; .npy / .wav → raw inference. Grayscale models and all-image multi-input models (super-resolution) are handled automatically. Full control for any model: model.infer({name: tensor}) with model.inputs.

Results holds
r.boxes xyxy, xywh, conf, cls
r.masks / r.keypoints / r.probs masks · [x,y,conf] · top1/top5
r.text decoded text (OCR, face attributes)
r.tensors raw {name: ndarray}
r.plot() / r.save(path) annotated image (or the model's output image)
CLI 
ovkit run detect image.jpg --save out.jpg   # one-shot inference
ovkit run age_gender face.jpg --device NPU
ovkit list                                  # aliases + models with descriptions
ovkit info face_detection                   # source / task / license
ovkit download detect                       # warm the cache
ovkit devices                               # available OpenVINO devices
GenAI (LLM / speech-to-text) 
from ovkit.genai import pipeline

llm = pipeline("llm")                        # tinyllama_chat from the mirror
print(llm.generate("Explain OpenVINO in one sentence.", max_new_tokens=64))

stt = pipeline("stt")                        # whisper_base
print(stt.generate(audio_16k_mono_float32))

Needs pip install "ovkit[genai]".

Web demo (image / webcam / audio / text) 
pip install -r examples/requirements.txt
python examples/web_app.py                   # http://127.0.0.1:8000

Pick any model — the right input (upload / webcam / audio / text) appears automatically and results render with overlays.

Adding a model

Models are data, not code — one manifest entry (src/ovkit/manifests/):

my_model:
  src: hf
  repo: leeyunjai/ovkit-models
  filename: detect/my_model/model.xml
  task: detect
  description: Shown by `ovkit list`.
  license: apache-2.0            # must be permissive — enforced at load time

Resolution: alias → local path → cache (~/.cache/ovkit) → download → convert → cache, with atomic writes, sha256 checks, upstream fallback, and OVKIT_OFFLINE=1. Maintainer tooling (mirror build / verify / self-check / benchmark) lives in scripts/ — see the guide.

License

ovkit is Apache-2.0 and license-clean by design: only permissive (Apache/MIT/BSD) models and libraries — no AGPL model stacks, no non-commercial weights; every manifest entry must declare a permissive license (enforced at load time).

Project links

Data verified by PyPI on Aug 23, 2026
Data provided by the project maintainers, verified at the time the release was uploaded to PyPI.

Key dates

PyPI data
Data sourced directly from PyPI's database.
  • Released:
Newer release available (0.2.0)

1 maintainer

PyPI data
Data sourced directly from PyPI's database.
Avatar for leeyunjai from gravatar.com leeyunjai

Credits

Author: ovkit contributors

GitHub Statistics

Data verified by PyPI on Aug 23, 2026
The GitHub source repository was provided by the project maintainers and verified by PyPI at the time of upload. Stars, forks, and open issues/PRs are derived from that repository and have not been independently verified.

License

Apache Software License (Apache-2.0)

Requires

Python >=3.10

Provides Extra

all anomaly dev genai quant

Tags

computer-vision detection inference npu openvino quantization

Classifiers

Report project as malware

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

ovkit-0.1.2.tar.gz (52.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

ovkit-0.1.2-py3-none-any.whl (66.5 kB view details)

Uploaded Python 3

File details

Details for the file ovkit-0.1.2.tar.gz.

File metadata

  • Download URL: ovkit-0.1.2.tar.gz
  • Upload date:
  • Size: 52.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for ovkit-0.1.2.tar.gz
Algorithm Hash digest
SHA256 cfc0870559c0ac05bb8e5942d85a8839ce71de2d52d46b57279f7fac880a4330
MD5 18545849524e0a5b6c1881892287bb8f
BLAKE2b-256 69d363e508d7ef1ef52e34d8a1b5be8b2cd2bdc14740a64c8cbde37f2dd6151a

See more details on using hashes here.

Provenance

The following attestation bundles were made for ovkit-0.1.2.tar.gz:

Publisher: release.yml on leeyunjai82/ovkit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file ovkit-0.1.2-py3-none-any.whl.

File metadata

  • Download URL: ovkit-0.1.2-py3-none-any.whl
  • Upload date:
  • Size: 66.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for ovkit-0.1.2-py3-none-any.whl
Algorithm Hash digest
SHA256 9b34bd8733a89656ade0a62aebca83724dffcb1813501776fab86da224df5fe2
MD5 d788539ae408ed3b0f92874e0bd5224c
BLAKE2b-256 d454adaa18db834fc153a7b14a4dae93acbba1ce7db20fa181348518486937c3

See more details on using hashes here.

Provenance

The following attestation bundles were made for ovkit-0.1.2-py3-none-any.whl:

Publisher: release.yml on leeyunjai82/ovkit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Release history Release notifications | RSS feed

0.2.0

2 files

This release

0.1.2 This release

2 files

0.1.1

2 files

0.1.0

2 files

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page