Source: https://seerepo.com/compare/NVIDIA/TensorRT-LLM/vs/ray-project/ray
Markdown source: https://seerepo.com/compare/NVIDIA/TensorRT-LLM/vs/ray-project/ray.md
Description: TensorRT-LLM and ray share 35 technologies: FastAPI, h5py, pyzmq, anyio, uvicorn, starlette, Rich, matplotlib, Plotly, accelerate, rouge-score, sentencepiece, tiktoken, torch, triton, transformers, diffusers, einops, safetensors, torchvision, datasets, evaluate, lm-eval, peft, jsonlines, pillow, jinja2, mako, Pydantic, jsonschema, opentelemetry-api, psutil, prometheus-client, boto3, and OpenTelemetry. Both are written in Python.

# TensorRT-LLM vs ray

TensorRT-LLM and ray share 35 technologies: FastAPI, h5py, pyzmq, anyio, uvicorn, starlette, Rich, matplotlib, Plotly, accelerate, rouge-score, sentencepiece, tiktoken, torch, triton, transformers, diffusers, einops, safetensors, torchvision, datasets, evaluate, lm-eval, peft, jsonlines, pillow, jinja2, mako, Pydantic, jsonschema, opentelemetry-api, psutil, prometheus-client, boto3, and OpenTelemetry. Both are written in Python.

- [NVIDIA/TensorRT-LLM](https://seerepo.com/r/NVIDIA/TensorRT-LLM): TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in a performant way.
- [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: FastAPI, h5py, pyzmq, anyio, uvicorn, starlette, Rich, matplotlib, Plotly, accelerate, rouge-score, sentencepiece, tiktoken, torch, triton, transformers, diffusers, einops, safetensors, torchvision, datasets, evaluate, lm-eval, peft, jsonlines, pillow, jinja2, mako, Pydantic, jsonschema, opentelemetry-api, psutil, prometheus-client, boto3, OpenTelemetry
- Only in TensorRT-LLM: blobfile, OpenAI, apache-tvm-ffi, cuda-python, flash-linear-attention, flashinfer-python, NVIDIA, nvidia-cudnn-frontend, openai-codex, openai-harmony, optimum, quack-kernels, rouge, torch-c-dlpack-ext, torchao, flash-attn, xgrammar, librosa, MCP, mistral-common, partial-json-parser, tritonclient, llguidance, msgspec, opentelemetry-semantic-conventions-ai, opentelemetry-sdk, prometheus-fastapi-instrumentator
- Only in ray: amqp, async-lru, azure-common, azure-datalake-store, azure-storage-blob, confluent-kafka, diskcache, expiringdict, gitdb, google-cloud-core, kombu, pickleshare, pylance, pymongo, pyarrow, aioboto3, MongoDB, Redis, adlfs, smart-open, pyiceberg, pymysql, clickhouse-connect, deltalake, gcsfs, Celery, fastavro, s3fs, webdataset, Google, Hugging Face, Google Cloud Storage, Kafka, asgiref, Flask, flask-cors, graphene, graphql-core, graphql-relay, jupyter-core, jupyterlab-server, werkzeug, Ray, AIOHTTP, Lightning, tornado, React, react-router-dom, alabaster, cycler, fonttools, jupyterlab-pygments, pydata-sphinx-theme, soupsieve, sphinx-design, tinycss2, contourpy, glfw, jupyter-server-terminals, jupyterlab-widgets, matplotlib-inline, prompt-toolkit, pyopengl, Textual, widgetsnbextension, xyzservices, Gradio, ipywidgets, pydot, alembic, peewee, sqlglot, SQLAlchemy, argon2-cffi, argon2-cffi-bindings, azure-identity, flask-basicauth, google-auth-oauthlib, itsdangerous, joserfc, jwcrypto, keyring, msal, msal-extensions, oauth2client, oauthlib, pyjwt, requests-oauthlib, cryptography, google-auth-httplib2, albucore, bitsandbytes, cmdstanpy, crcmod, cupy-cuda12x, cupy-cuda13x, fairscale, gradio-client, gym, LangChain, llvmlite, ml-dtypes, MLflow, nltk, ONNX, onnxruntime, onnxscript, openai-whisper, opt-einsum, orbax-checkpoint, prophet, sacrebleu, scikit-image, skops, tf-keras, timm, tokenizers, torchtext, gymnasium, TensorFlow, jax, xgboost, lightgbm, scikit-learn, pytorch-lightning, flax, Keras, torchmetrics, albumentations, deepspeed, hyperopt, Optuna, pettingzoo, tensordict, tensorflow-datasets, tensorflow-probability, torchaudio, bleach, defusedxml, et-xmlfile, imageio, imageio-ffmpeg, imagesize, linkify-it-py, lxml, markdown, markdown-it-py, markupsafe, mdurl, mistune, moviepy, nbconvert, nh3, openpyxl, pandocfilters, ruamel-yaml, webencodings, xlrd, xmltodict, docutils, beautifulsoup4, av, nbformat, torchcodec, annotated-types, fastjsonschema, fqdn, jsonschema-specifications, pathvalidate, pydantic-core, referencing, rfc3339-validator, rfc3986, rfc3986-validator, validators, openapi-schema-validator, aiobotocore, azure-mgmt-compute, azure-mgmt-resource, colorlog, Gunicorn, opentelemetry-instrumentation, opentelemetry-semantic-conventions, python-json-logger, botocore, kubernetes, opentelemetry-exporter-prometheus, docker, opentelemetry-exporter-otlp-proto-grpc, opentelemetry-instrumentation-fastapi, typeface-roboto, MUI, usehooks-ts, lowlight, react-window, react-icons, @mui/icons-material, js-yaml

## Side by side

| | TensorRT-LLM | ray |
|---|---|---|
| Size | 2,884,778 lines written, 3,120 code files | 1,531,664 lines written, 2,777 code files |
| Who writes it | 1.7% of its commits signed by AI. With Claude Code, CodeRabbit, GitHub Copilot, and Cursor | 3.6% of its commits signed by AI. With Claude Code, Gemini CLI, GitHub Copilot, Cursor, and Codex |
| Activity | Last push November 8, 2026, 10,997 commits, 15,038 stars | Last push November 6, 2026, 32,373 commits, 44,213 stars |
| Tests | 2,703 test files. 58.6% of its code files have a test that imports them | 2,168 test files. 9.7% 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, TensorRT-LLM or ray?

TensorRT-LLM, with 2,884,778 lines written; ray has 1,531,664.

The proof: [NVIDIA/TensorRT-LLM's code at commit b433473](https://github.com/NVIDIA/TensorRT-LLM/tree/b433473ba0d0d49764f4d2fb3d45c82716b946dc)

### Which has more GitHub stars, TensorRT-LLM or ray?

ray, with 44,213 stars; TensorRT-LLM has 15,038.

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

### Which has more tests, TensorRT-LLM or ray?

TensorRT-LLM, with 2,703 test files; ray has 2,168.

The proof: [NVIDIA/TensorRT-LLM's code at commit b433473](https://github.com/NVIDIA/TensorRT-LLM/tree/b433473ba0d0d49764f4d2fb3d45c82716b946dc)

### Which was updated more recently, TensorRT-LLM or ray?

TensorRT-LLM: its last push was on November 8, 2026, and ray's on November 6, 2026.

The proof: [NVIDIA/TensorRT-LLM's commits on GitHub](https://github.com/NVIDIA/TensorRT-LLM/commits)

### Which has more commits signed by AI, TensorRT-LLM or ray?

ray: an AI signed 3.6% of its commits, with Claude Code, Gemini CLI, GitHub Copilot, Cursor, and Codex; in TensorRT-LLM, 1.7%.

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

## Sources

- NVIDIA/TensorRT-LLM: Read on November 8, 2026, at commit b433473 (https://github.com/NVIDIA/TensorRT-LLM/tree/b433473ba0d0d49764f4d2fb3d45c82716b946dc)
- ray-project/ray: Read on November 6, 2026, at commit 77567c5 (https://github.com/ray-project/ray/tree/77567c5c5790e9e91aa40a5af93004ef958c2ffc)

## 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.