bentoml has 213K monthly downloads and 9K GitHub stars. Version 1.4.39 supports Python 3.9+, first published in 2019.
It is stably maintained under a permissive license, last released in May 2026.
It is built on top of a2wsgi, aiohttp and aiohttp-asgi-connector. It is used by 24 tracked packages, including nubison-model, konfuzio-sdk and openllm.
GitHub topics include model-serving, mlops and llmops.
Package Insights
((week_daily_avg - month_daily_avg) / month_daily_avg) * 100Weekly Downloads
GitHub Stars
Downloads by OS
Python Versions
Top Countries
Dependencies
- a2wsgi >=1.10.7
- aiohttp
- aiohttp-asgi-connector >=1.1.2
- aiosqlite >=0.20.0
- attrs >=22.2.0
- cattrs <23.2.0,>=22.1.0
- click >=7.0
- click-option-group
- cloudpickle >=2.0.0
- fsspec >=2025.7.0
- httpx
- httpx-ws >=0.6.0
- jinja2 >=3.0.1
- kantoku >=0.18.3
- numpy
- nvidia-ml-py
- opentelemetry-api ~=1.20
- opentelemetry-instrumentation ~=0.41b0
- opentelemetry-instrumentation-aiohttp-client ~=0.41b0
- opentelemetry-instrumentation-asgi ~=0.41b0
- opentelemetry-sdk ~=1.20
- opentelemetry-semantic-conventions ~=0.41b0
- opentelemetry-util-http ~=0.41b0
- packaging >=22.0
- pathspec
- pip-requirements-parser >=31.2.0
- prometheus-client >=0.10.0
- psutil
- pydantic <3
- python-dateutil
- python-json-logger
- python-multipart
- pyyaml >=5.0
- rich >=11.2.0
- rich-toolkit >=0.15.1
- schema
- simple-di >=0.1.4
- starlette >=0.24.0
- tomli >=1.1.0
- tomli-w
- uvicorn >=0.22.0
- watchfiles >=0.15.0
46 optional dependencies
- bentoml-unsloth[unsloth]
- grpcio[all]
- grpcio[grpc]
- grpcio[grpc-channelz]
- grpcio[grpc-reflection]
- grpcio-channelz[all]
- grpcio-channelz[grpc-channelz]
- grpcio-health-checking[grpc-reflection]
- grpcio-health-checking[all]
- grpcio-health-checking[grpc]
- grpcio-health-checking[grpc-channelz]
- grpcio-reflection[all]
- grpcio-reflection[grpc-reflection]
- opentelemetry-exporter-jaeger[tracing]
- opentelemetry-exporter-jaeger[all]
- opentelemetry-exporter-jaeger[tracing-jaeger]
- opentelemetry-exporter-otlp[tracing]
- opentelemetry-exporter-otlp[all]
- opentelemetry-exporter-otlp[tracing-otlp]
- opentelemetry-exporter-otlp-proto-grpc[all]
- opentelemetry-exporter-otlp-proto-grpc[monitor-otlp]
- opentelemetry-exporter-otlp-proto-http[monitor-otlp]
- opentelemetry-exporter-otlp-proto-http[all]
- opentelemetry-exporter-zipkin[tracing]
- opentelemetry-exporter-zipkin[all]
- opentelemetry-exporter-zipkin[tracing-zipkin]
- opentelemetry-instrumentation-grpc[grpc-channelz]
- opentelemetry-instrumentation-grpc[grpc]
- opentelemetry-instrumentation-grpc[all]
- opentelemetry-instrumentation-grpc[grpc-reflection]
- pandas[all]
- pandas[io-pandas]
- pandas[io]
- pillow[io-image]
- pillow[io]
- pillow[all]
- protobuf[grpc-channelz]
- protobuf[grpc]
- protobuf[all]
- protobuf[grpc-reflection]
- pyarrow[io]
- pyarrow[all]
- pyarrow[io-pandas]
- s3fs[all]
- s3fs[aws]
- tritonclient[triton]
Used By
- nubison-model
- konfuzio-sdk
- openllm
- clip-api-service
- sentencebertservice
- comfy-pack
- bento2seldon
- bento-sgl-router
- bentoctl
- dreamml
- bentoml-unsloth
- llmchatbot
- ml-assistant-cli
- onediffusion
- bentomlx
- raptor-labsdk
- aiserve
- bentoml-plugins-arize
- fusionmllib
- kazemlstack
- nidam
- moark-comfyui-engine
- bentoml-comfyui
- modelo-proyeccion-subastas-autos