sybrenjansen/mpire

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

2064 stars · 40 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

MPIRE (MultiProcessing Is Really Easy) is a Python library providing a faster, more feature-rich alternative to the standard multiprocessing package. It offers pool-based map/apply functions with copy-on-write shared objects, per-worker state, progress bars, timeouts, and worker insights.

Use cases

• parallelize map operations across CPU cores in Python
• load a large model once per worker process
• show a progress bar for multiprocessing tasks
• run parallel tasks with timeouts and graceful exception handling
• parallelize lambdas and functions in Jupyter notebooks
• reduce memory usage when running many parallel tasks

When to choose

• you need multiprocessing that is faster and easier than Python's multiprocessing.Pool
• you want per-worker state, init/exit hooks, or copy-on-write shared objects
• you want built-in progress bars, dashboards, and worker efficiency insights
• you need to parallelize exotic objects or lambdas via dill

When to avoid

• you need distributed or multi-machine parallelism rather than a single machine
• you rely on the spawn start method and need copy-on-write sharing
• your workload is I/O-bound and better served by asyncio or threading

Facets

library · maturity active

concurrency developer-tools developer-tools performance cross-platform python windows multiprocessing parallel-processing worker-pool progress-bar tqdm dill cpu-pinning linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="sybrenjansen/mpire")