# ShannonAI/service-streamer

Boosting your Web Services of Deep Learning Applications.

Repository: https://github.com/ShannonAI/service-streamer
Canonical: https://ross.abutalabs.com/products/service-streamer
Language: Python
License: Apache-2.0
License Family: permissive
Topics: deep-learning, web, pytorch, model-deployment, tensorflow, bert
Last push: 2021-05-13T02:07:19+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2585, "days_push": 1939, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1241, forks 187 (observed 2026-08-28T04:04:06.305545+00:00)

## What it is
Service Streamer is a Python middleware that queues discrete web service requests into mini-batches for deep learning model inference, improving GPU utilization. It works with any web framework and deep learning framework, including multi-GPU setups.

## Use cases
- serve BERT inference behind a web API with high throughput
- batch concurrent user requests into mini-batches for GPU inference
- deploy a PyTorch or TensorFlow model as a low-latency online service
- scale model inference across multiple GPUs
- reduce latency of deep learning web services under concurrent load

## When to choose
- you serve a deep learning model via a web service and GPUs are underutilized by one-request-at-a-time handling
- you want dynamic batching with minimal code changes
- you need multi-GPU request distribution for inference

## When to avoid
- your model inference is CPU-bound or not batch-friendly
- you need a full-featured model server with versioning and monitoring
- you require actively maintained software with recent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: middleware, machine-learning, llm-inference, streaming, concurrency
- domain: deep-learning, machine-learning, web-development, backend
- platform: python
- tags: model-serving, dynamic-batching, gpu-utilization, mini-batching, inference-serving, natural-language-processing, linux, docker

## Member repositories
- ShannonAI/service-streamer (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:06.305545+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 2026-08-30T05:08:27.528799+00:00, confidence not recorded.
  - readme: https://github.com/ShannonAI/service-streamer (fetched 2026-08-28T04:04:06.305545+00:00, sha 0f797c47ef76)
- Data as of 2026-08-30T08:39:29.467469+00:00.