# NervanaSystems/maxas

Assembler for NVIDIA Maxwell architecture

Repository: https://github.com/NervanaSystems/maxas
Canonical: https://ross.abutalabs.com/products/maxas
Homepage: https://github.com/NervanaSystems/maxas
Language: Sass
License: MIT
License Family: permissive
Archived: true
Last push: 2023-01-03T22:54:15+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 4245, "days_push": 1338, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1074, forks 174 (observed 2026-08-28T04:03:29.102555+00:00)

## What it is
MaxAs is an assembler for NVIDIA Maxwell GPU architecture that assembles SASS (Shader Assembly) instructions, written in Perl. It enables low-level GPU kernel authoring, including a hand-tuned SGEMM example, and is now officially discontinued by Intel.

## Use cases
- assemble SASS instructions for NVIDIA Maxwell GPUs
- write hand-optimized GPU kernels in shader assembly
- study Maxwell instruction scheduling and control codes
- build a highly optimized SGEMM kernel from scratch
- reverse-engineer or patch compiled CUDA kernel assembly

## When to choose
- you need precise control over Maxwell SASS instruction encoding or scheduling
- you are researching GPU microarchitecture or instruction-level optimization on Maxwell hardware

## When to avoid
- you target newer NVIDIA architectures (Pascal, Volta, Ampere, etc.)
- you want a maintained tool - the project is archived and no longer accepts patches
- you prefer writing kernels in CUDA C rather than raw assembly

## Facets
- artifact type: library
- maturity: abandoned
- function: compiler, developer-tools, gpu-computing
- domain: gpu-computing, compilers, hardware, deep-learning
- platform: cli, python
- tags: sass-assembly, nvidia-maxwell, cuda, assembler, perl, discontinued, linux

## Member repositories
- NervanaSystems/maxas (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:29.102555+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 2026-08-30T06:53:30.102143+00:00, confidence not recorded.
  - readme: https://github.com/NervanaSystems/maxas (fetched 2026-08-28T04:03:29.102555+00:00, sha 133403aa6c75)
  - homepage: https://github.com/NervanaSystems/maxas (fetched 2026-08-29T12:55:04.696055+00:00, sha d52e27c75d59)
- Data as of 2026-08-30T08:39:29.467469+00:00.