Metadata-Version: 2.4
Name: vllm-sr-sim
Version: 0.1.0.dev20260318114607
Summary: vLLM Semantic Router fleet simulator for capacity planning, SLO validation, and what-if analysis
Author: vLLM Semantic Router Team
License: Apache-2.0
Project-URL: Homepage, https://github.com/vllm-project/semantic-router
Project-URL: Documentation, https://vllm-semantic-router.com/docs/fleet-sim/overview
Project-URL: Repository, https://github.com/vllm-project/semantic-router/tree/main/src/fleet-sim
Project-URL: Issues, https://github.com/vllm-project/semantic-router/issues
Keywords: vllm,semantic-router,simulation,capacity-planning,gpu
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Requires-Python: >=3.10
Description-Content-Type: text/markdown
Requires-Dist: numpy>=1.24
Requires-Dist: scipy>=1.11
Provides-Extra: api
Requires-Dist: fastapi>=0.110; extra == "api"
Requires-Dist: uvicorn[standard]>=0.27; extra == "api"
Requires-Dist: python-multipart>=0.0.9; extra == "api"
Provides-Extra: dev
Requires-Dist: pytest>=7; extra == "dev"
Requires-Dist: black; extra == "dev"
Requires-Dist: ruff; extra == "dev"
Requires-Dist: fastapi>=0.110; extra == "dev"
Requires-Dist: uvicorn[standard]>=0.27; extra == "dev"
Requires-Dist: python-multipart>=0.0.9; extra == "dev"
Requires-Dist: httpx>=0.27; extra == "dev"

# vllm-sr-sim

`vllm-sr-sim` is the maintained fleet simulator for this repository. It sizes heterogeneous GPU fleets, evaluates routing strategies, and exposes a service mode that the dashboard can call across containers.

Repository-maintained docs now live in the website:

- https://vllm-semantic-router.com/docs/fleet-sim/overview
- https://vllm-semantic-router.com/docs/fleet-sim/getting-started
- https://vllm-semantic-router.com/docs/fleet-sim/use-cases

## Install

```bash
cd src/fleet-sim
pip install -e .
```

Install the service extras when you want to run the simulator API:

```bash
pip install -e .[api]
```

For local development and tests:

```bash
pip install -e .[dev]
```

## CLI

```bash
vllm-sr-sim --version

vllm-sr-sim optimize \
  --cdf data/azure_cdf.json \
  --lam 200 --slo 500 --b-short 6144 \
  --verify-top 3 --n-sim-req 30000

vllm-sr-sim whatif \
  --cdf data/azure_cdf.json \
  --lam-range 50 100 200 500 1000 \
  --slo 500 --b-short 6144

vllm-sr-sim serve --host 0.0.0.0 --port 8000
```

`vllm-sr serve` also starts `vllm-sr-sim` by default as a sibling container on the shared runtime network so the dashboard can proxy it without rebuilding the router image.

## Layout

- `fleet_sim/`: simulation engine, optimizers, routing, hardware, workload, and service package
- `run_sim.py`: unified CLI entrypoint used by `vllm-sr-sim`
- `tests/`: simulator and service test coverage
- `data/`: reference workload traces used by the examples and dashboard integration
- `examples/`: sample scripts and multi-pool input files

## Docs

Long-form simulator docs are maintained in the repository website. Keep the package README focused on installation, CLI usage, and source layout.
