Source: https://seerepo.com/compare/OpenDCAI/DataFlow/vs/ray-project/ray
Markdown source: https://seerepo.com/compare/OpenDCAI/DataFlow/vs/ray-project/ray.md
Description: DataFlow and ray share 18 technologies: gcsfs, pyarrow, Google, FastAPI, uvicorn, Gradio, Rich, sqlglot, sentencepiece, torchvision, torch, transformers, nltk, datasets, tiktoken, Pydantic, psutil, and colorlog. Both are written in Python.

# DataFlow vs ray

DataFlow and ray share 18 technologies: gcsfs, pyarrow, Google, FastAPI, uvicorn, Gradio, Rich, sqlglot, sentencepiece, torchvision, torch, transformers, nltk, datasets, tiktoken, Pydantic, psutil, and colorlog. Both are written in Python.

- [OpenDCAI/DataFlow](https://seerepo.com/r/OpenDCAI/DataFlow): [SIGMOD'27] Easy Data Preparation with latest LLMs-based Operators and Pipelines.
- [ray-project/ray](https://seerepo.com/r/ray-project/ray): Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

## What they share and what sets them apart

- In common: gcsfs, pyarrow, Google, FastAPI, uvicorn, Gradio, Rich, sqlglot, sentencepiece, torchvision, torch, transformers, nltk, datasets, tiktoken, Pydantic, psutil, colorlog
- Only in DataFlow: db-dtypes, pymysql, OpenAI, accelerate, evaluate, torchaudio, chonkie, fasttext, librosa, math-verify, ModelScope, presidio-analyzer, presidio-anonymizer, lxml-html-clean, pypdf, pymupdf
- Only in ray: aimrecords, aimrocks, amqp, array-record, arro3-core, async-lru, azure-common, azure-datalake-store, azure-storage-blob, confluent-kafka, delta-kernel-rust-sharing-wrapper, diskcache, expiringdict, feather-format, gitdb, google-cloud-core, h5py, huey, jupyter-cache, kombu, partd, petastorm, pickleshare, py-partiql-parser, pylance, pymongo, qpd, s3torchconnectorclient, s3transfer, semidbm, sqlitedict, taskiq, tensorboard-data-server, zict, mcap, aioboto3, MongoDB, Redis, adlfs, pyiceberg, smart-open, clickhouse-connect, daft, deltalake, lance-namespace, obstore, Celery, delta-sharing, fastavro, hudi, pyzmq, s3fs, s3torchconnector, taskiq-redis, zarr, Hugging Face, Google Cloud Storage, Kafka, Databricks, adagio, asgiref, Flask, flask-cors, graphene, graphql-core, graphql-relay, jupyter-core, jupyterlab-server, pygame, pymars, werkzeug, Ray, starlette, AIOHTTP, pyspark, Lightning, raydp, anyio, tornado, React, react-router-dom, accessible-pygments, alabaster, cycler, fonttools, jupyterlab-pygments, pydata-sphinx-theme, soupsieve, sphinx-design, tinycss2, aim-ui, alive-progress, bokeh, contourpy, glfw, halo, jupyter-server-terminals, jupyterlab-widgets, matplotlib-inline, Plotly, prompt-toolkit, pyglet, pyopengl, spinners, Textual, widgetsnbextension, xyzservices, ipywidgets, pydot, alembic, peewee, adal, argon2-cffi, argon2-cffi-bindings, azure-identity, flask-basicauth, gcs-oauth2-boto-plugin, google-auth-oauthlib, google-oauth, google-reauth, id, itsdangerous, joserfc, jwcrypto, keyring, msal, msal-extensions, oauth2client, oauthlib, pyjwt, pyu2f, requests-oauthlib, cryptography, google-auth-httplib2, aws-requests-auth, albucore, autograd, ax-platform, bayesian-optimization, bitsandbytes, botorch, cma, cmdstanpy, crcmod, cupy-cuda100, cupy-cuda12x, cupy-cuda13x, einops, fairscale, fbgemm-gpu, gpytorch, gradio-client, gym, jaxlib, labmaze, LangChain, lightning-utilities, linear-operator, llvmlite, lm-eval, minigrid, ml-dtypes, moocore, mosaicml, ONNX, onnx-ir, onnxruntime, onnxscript, openai-whisper, opt-einsum, optax, orbax-checkpoint, peft, prophet, pymoo, pyro-ppl, pytorch-ranger, rouge-score, sacrebleu, safetensors, scikit-image, skops, stanio, statsforecast, tf-keras, timm, tokenizers, torch-geometric, torch-optimizer, torch-scatter, torch-sparse, torchtext, triton, yahp, gymnasium, TensorFlow, jax, xgboost, lightgbm, scikit-learn, MLflow, pytorch-lightning, comet-ml, dm-env, flax, Keras, torchmetrics, aim, ale-py, configspace, deepspeed, diffusers, dm-control, hpbandster, hyperopt, nevergrad, open-spiel, Optuna, tensorflow-probability, zoopt, bleach, defusedxml, et-xmlfile, imageio, imageio-ffmpeg, imagesize, jsonlines, linkify-it-py, lxml, mako, markdown, markdown-it-py, markupsafe, mdit-py-plugins, mdurl, mistune, moviepy, myst-parser, nbconvert, nh3, openpyxl, pandocfilters, pytablewriter, readme-renderer, ruamel-yaml, ruamel-yaml-clib, strictyaml, tinyscaler, webencodings, xlrd, xmltodict, pillow, docutils, jinja2, beautifulsoup4, decord, nbformat, torchcodec, annotated-types, fastjsonschema, fqdn, jsonschema-path, jsonschema-specifications, openapi-spec-validator, pathvalidate, referencing, rfc3339-validator, rfc3986, rfc3986-validator, typepy, validators, jsonschema, aiobotocore, applicationinsights, aws-sam-translator, aws-xray-sdk, azure-cli-core, azure-cli-telemetry, azure-core, azure-mgmt-compute, azure-mgmt-core, azure-mgmt-network, azure-mgmt-resource, azure-mgmt-resource-deployments, boto, cloud-tpu-diagnostics, execnet, google-cloud-audit-log, Gunicorn, log-symbols, ml-goodput-measurement, msrestazure, opencensus-context, opentelemetry-instrumentation, opentelemetry-instrumentation-asgi, opentelemetry-semantic-conventions, opentelemetry-util-http, python-json-logger, supervisor, boto3, botocore, prometheus-client, pybuildkite, OpenTelemetry, kubernetes, opentelemetry-exporter-prometheus, docker, opentelemetry-exporter-otlp-proto-grpc, opentelemetry-instrumentation-fastapi, ray-haproxy, typeface-roboto, MUI, usehooks-ts, lowlight, react-window, react-icons, @mui/icons-material, js-yaml

## Side by side

| | DataFlow | ray |
|---|---|---|
| Size | 78,684 lines written, 317 code files | 1,566,356 lines written, 2,810 code files |
| Who writes it | 0.2% of its commits signed by AI. With Claude Code and GitHub Copilot | 3.7% of its commits signed by AI. With Claude Code, Gemini CLI, GitHub Copilot, Cursor, and Codex |
| Activity | Last push November 16, 2026, 999 commits, 8,607 stars | Last push November 13, 2026, 32,468 commits, 44,282 stars |
| Tests | 83 test files. 14.8% of its code files have a test that imports them | 2,208 test files. 9.4% of its code files have a test that imports them |

- Size: Lines written by people and agents, without generated ones; code files, without tests.
- Who writes it: A commit counts as AI's when it carries its signature: a Co-Authored-By line, its author name or the merge of its branch.
- Activity: From GitHub and its history, on the day SeeRepo read it.
- Tests: The files that test the code, and the share of code files a test imports.

## Frequently asked questions

### Which is bigger, DataFlow or ray?

ray, with 1,566,356 lines written; DataFlow has 78,684.

The proof: [ray-project/ray's code at commit de4c576](https://github.com/ray-project/ray/tree/de4c5765b6cf0265f43f9edccf5bd3821f3a9468)

### Which has more GitHub stars, DataFlow or ray?

ray, with 44,282 stars; DataFlow has 8,607.

The proof: [ray-project/ray on GitHub](https://github.com/ray-project/ray)

### Which has more tests, DataFlow or ray?

ray, with 2,208 test files; DataFlow has 83.

The proof: [ray-project/ray's code at commit de4c576](https://github.com/ray-project/ray/tree/de4c5765b6cf0265f43f9edccf5bd3821f3a9468)

### Which was updated more recently, DataFlow or ray?

DataFlow: its last push was on November 16, 2026, and ray's on November 13, 2026.

The proof: [OpenDCAI/DataFlow's commits on GitHub](https://github.com/OpenDCAI/DataFlow/commits)

### Which has more commits signed by AI, DataFlow or ray?

ray: an AI signed 3.7% of its commits, with Claude Code, Gemini CLI, GitHub Copilot, Cursor, and Codex; in DataFlow, 0.2%.

The proof: [the commits each tool signed in ray-project/ray](https://seerepo.com/r/ray-project/ray/ai)

## Sources

- OpenDCAI/DataFlow: Read on November 16, 2026, at commit 13d5de8 (https://github.com/OpenDCAI/DataFlow/tree/13d5de84ad151c0bfc434167d9e3b9e36c3809e4)
- ray-project/ray: Read on November 13, 2026, at commit de4c576 (https://github.com/ray-project/ray/tree/de4c5765b6cf0265f43f9edccf5bd3821f3a9468)

## What is your repo built with?

Any public repo, with no account: https://seerepo.com/r/owner/repo/built, with the owner and name it has on GitHub.
