# Bend

A massively parallel, high-level programming language

Repository: https://github.com/HigherOrderCO/Bend
Canonical: https://ross.abutalabs.com/products/bend
Homepage: https://higherorderco.com
Language: Rust
License: Apache-2.0
License Family: permissive
Last push: 2026-07-07T19:51:01+00:00
Link (homepage): https://higherorderco.com

## Health v2 (maintenance only)
Score: 69/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 91, release rhythm 35, longevity 78
- inputs: {"age_days": 1100, "days_push": 57, "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 19808, forks 494 (observed 2026-08-28T04:11:29.267973+00:00)

## What it is
Bend is a high-level, massively parallel programming language that feels like Python or Haskell but automatically parallelizes programs across multi-core CPUs and NVIDIA GPUs without explicit concurrency annotations. It is powered by HVM2, an interaction-combinator-based runtime that achieves near-linear speedup with core count.

## Use cases
- write parallel code without locks, mutexes, or atomics
- run programs massively parallel on NVIDIA GPUs
- parallelize recursive and higher-order functional programs
- achieve near-linear speedup on multi-core CPUs
- compile Bend programs to standalone C or CUDA for maximum performance
- experiment with interaction net computation models

## When to choose
- you want implicit, automatic parallelism on CPUs or NVIDIA GPUs
- your workload scales with core count and single-core performance is less critical
- you prefer a Python/Haskell-like high-level language over writing CUDA
- you want to explore interaction combinator-based computation

## When to avoid
- you need Windows support natively (only WSL2 works)
- you need non-NVIDIA GPU support
- you need mature, optimized single-core performance comparable to GCC or GHC
- you need a production-stable language toolchain

## Facets
- artifact type: cli-tool
- maturity: experimental
- function: compiler, interpreter, concurrency, gpu-computing
- domain: programming-languages, compilers, gpu-computing, developer-tools
- platform: cli, rust
- tags: interaction-combinators, functional-programming, implicit-parallelism, cuda, hvm2, high-level-language, programming-language, linux, macos, gpu

## Member repositories
- HigherOrderCO/Bend (main) score 69
- HigherOrderCO/HVM2 (backend) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:29.267973+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 2026-08-30T08:30:58.856111+00:00, confidence not recorded.
  - readme: https://github.com/HigherOrderCO/Bend (fetched 2026-08-28T04:11:29.267973+00:00, sha 34af29b16768)
  - homepage: https://higherorderco.com (fetched 2026-08-29T07:58:04.264676+00:00, sha 79cef861f61c)
- Data as of 2026-08-30T08:39:29.467469+00:00.