Flash

pip install runpod-flash                 # requires Python >=3.10

# auth option 1: browser-based login (saves token locally)
flash login

# auth option 2: API key via environment variable
export RUNPOD_API_KEY=your_key

flash init my-project                    # scaffold a new project in ./my-project

flash run                                # start local dev server at localhost:8888
flash run --auto-provision               # same, but pre-provision endpoints (no cold start)
flash build                              # package artifact for deployment (500MB limit)
flash build --exclude pkg1,pkg2          # exclude packages from build
flash deploy                             # build + deploy (auto-selects env if only one)
flash deploy --env staging               # build + deploy to "staging" environment
flash deploy --app my-app --env prod     # deploy a specific app to an environment
flash deploy --preview                   # build + launch local preview in Docker
flash env list                           # list deployment environments
flash env create staging                 # create "staging" environment
flash env get staging                    # show environment details + resources
flash env delete staging                 # delete environment + tear down resources
flash undeploy list                      # list all active endpoints
flash undeploy my-endpoint               # remove a specific endpoint

from runpod_flash import Endpoint, GpuGroup

@Endpoint(name="my-worker", gpu=GpuGroup.AMPERE_80, workers=5, dependencies=["torch"])
async def compute(data):
    import torch  # MUST import inside function (cloudpickle)
    return {"sum": torch.tensor(data, device="cuda").sum().item()}

result = await compute([1, 2, 3])

from runpod_flash import Endpoint, GpuGroup

api = Endpoint(name="my-api", gpu=GpuGroup.ADA_24, workers=(1, 5), dependencies=["torch"])

@api.post("/predict")
async def predict(data: list[float]):
    import torch
    return {"result": torch.tensor(data, device="cuda").sum().item()}

@api.get("/health")
async def health():
    return {"status": "ok"}

from runpod_flash import Endpoint, GpuGroup, PodTemplate

server = Endpoint(
    name="my-server",
    image="my-org/my-image:latest",
    gpu=GpuGroup.AMPERE_80,
    workers=1,
    env={"HF_TOKEN": "xxx"},
    template=PodTemplate(containerDiskInGb=100),
)

# LB-style
result = await server.post("/v1/completions", {"prompt": "hello"})
models = await server.get("/v1/models")

# QB-style
job = await server.run({"prompt": "hello"})
await job.wait()
print(job.output)

ep = Endpoint(id="abc123")
job = await ep.runsync({"input": "hello"})
print(job.output)

Endpoint(
    name="endpoint-name",                  # required (unless id= set)
    id=None,                               # connect to existing endpoint
    gpu=GpuGroup.AMPERE_80,               # single GPU type (default: ANY)
    gpu=[GpuGroup.ADA_24, GpuGroup.AMPERE_80],  # or list for auto-select by supply
    cpu=CpuInstanceType.CPU5C_4_8,        # CPU type (mutually exclusive with gpu)
    workers=5,                             # shorthand for (0, 5)
    workers=(1, 5),                        # explicit (min, max)
    idle_timeout=60,                       # seconds before scale-down (default: 60)
    dependencies=["torch"],                # pip packages for remote exec
    system_dependencies=["ffmpeg"],        # apt-get packages
    image="org/image:tag",                 # pre-built Docker image (client mode)
    env={"KEY": "val"},                    # environment variables
    volume=NetworkVolume(...),             # persistent storage
    gpu_count=1,                           # GPUs per worker
    template=PodTemplate(containerDiskInGb=100),
    flashboot=True,                        # fast cold starts
    execution_timeout_ms=0,                # max execution time (0 = unlimited)
)

NetworkVolume(name="my-vol", size=100)  # size in GB, default 100

PodTemplate(
    containerDiskInGb=64,    # container disk size (default 64)
    dockerArgs="",           # extra docker arguments
    ports="",                # exposed ports
    startScript="",          # script to run on start
)

job = await ep.run({"data": [1, 2, 3]})
await job.wait(timeout=120)        # poll until done
print(job.id, job.output, job.error, job.done)
await job.cancel()