# weld-project/weld

High-performance runtime for data analytics applications

Repository: https://github.com/weld-project/weld
Canonical: https://ross.abutalabs.com/products/weld
Homepage: https://www.weld.rs
Language: Rust
License: BSD-3-Clause
License Family: permissive
Topics: stanford, data, analytics, machine-learning, code-generation, performance, rust, llvm, pandas
Last push: 2026-04-13T17:23:52+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 77, release rhythm 8, longevity 100
- inputs: {"age_days": 3733, "days_push": 142, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3006, forks 252 (observed 2026-08-28T04:07:37.633118+00:00)

## What it is
Weld is a language and high-performance runtime that optimizes data-intensive analytics workflows by expressing computations in a common intermediate representation and compiling them with LLVM. It lets libraries like Pandas (via Grizzly) defer and jointly optimize across framework boundaries instead of moving data between isolated functions.

## Use cases
- speed up pandas workflows that are bottlenecked by data movement
- compile and optimize multi-library analytics pipelines end to end
- accelerate machine learning preprocessing in Python
- build a JIT compiler for data analytics code
- reduce overhead when chaining NumPy/Pandas operations

## When to choose
- your analytics pipeline spends most time moving data between library calls
- you want LLVM-based code generation for data-intensive workloads
- you need cross-library optimization of Pandas-like computations

## When to avoid
- you need a project with frequent updates and modern Rust/LLVM support
- your workload is small enough that data movement overhead is negligible
- you need production support for LLVM versions newer than 6.0

## Facets
- artifact type: library
- maturity: maintenance
- function: compiler, interpreter, machine-learning, data-science, benchmarking
- domain: data-science, analytics, performance, compilers, machine-learning
- platform: rust, cpp, python
- tags: llvm, intermediate-representation, pandas, code-generation, jit-compilation, data-analytics, macos, linux

## Member repositories
- weld-project/weld (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:37.633118+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 2026-08-30T07:30:22.906993+00:00, confidence not recorded.
  - readme: https://github.com/weld-project/weld (fetched 2026-08-28T04:07:37.633118+00:00, sha 17336206c8a2)
  - homepage: https://www.weld.rs (fetched 2026-08-29T09:45:37.326470+00:00, sha 05521d5998cf)
  - registry_crates: https://crates.io/api/v1/crates/weld (fetched 2026-08-29T09:45:37.335276+00:00, sha 7d5c873b9882)
- Data as of 2026-08-30T08:39:29.467469+00:00.