{"id":22054776,"url":"https://github.com/tgale96/grouped_gemm","last_synced_at":"2026-01-18T00:26:24.985Z","repository":{"id":195896790,"uuid":"693835633","full_name":"tgale96/grouped_gemm","owner":"tgale96","description":"PyTorch bindings for CUTLASS grouped GEMM.","archived":false,"fork":false,"pushed_at":"2024-10-31T19:40:59.000Z","size":44,"stargazers_count":84,"open_issues_count":12,"forks_count":57,"subscribers_count":4,"default_branch":"main","last_synced_at":"2025-04-29T06:16:57.359Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Cuda","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/tgale96.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2023-09-19T20:08:53.000Z","updated_at":"2025-04-25T17:52:03.000Z","dependencies_parsed_at":"2024-08-26T16:41:03.757Z","dependency_job_id":"0e8ad8ee-3b61-41d5-8e4e-e239837d2db6","html_url":"https://github.com/tgale96/grouped_gemm","commit_stats":null,"previous_names":["tgale96/grouped_gemm"],"tags_count":5,"template":false,"template_full_name":null,"purl":"pkg:github/tgale96/grouped_gemm","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tgale96%2Fgrouped_gemm","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tgale96%2Fgrouped_gemm/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tgale96%2Fgrouped_gemm/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tgale96%2Fgrouped_gemm/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tgale96","download_url":"https://codeload.github.com/tgale96/grouped_gemm/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tgale96%2Fgrouped_gemm/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28523717,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-17T23:53:28.710Z","status":"ssl_error","status_checked_at":"2026-01-17T23:52:20.131Z","response_time":85,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-30T16:01:43.351Z","updated_at":"2026-01-18T00:26:24.972Z","avatar_url":"https://github.com/tgale96.png","language":"Cuda","readme":"# Grouped GEMM\n\nA lighweight library exposing grouped GEMM kernels in PyTorch.\n\n# Installation\n\nRun `pip install grouped_gemm` to install the package.\n\n# Compiling from source\n\nBy default, the installed package runs in conservative (`cuBLAS`) mode:\nit launches one GEMM kernel per batch element instead of using a single\ngrouped GEMM kernel for the whole batch.\n\nTo enable using grouped GEMM kernels, you need to switch to the `CUTLASS`\nmode by setting the `GROUPED_GEMM_CUTLASS` environment variable to `1`\nwhen building the library. For example, to build the library in `CUTLASS`\nmode for Ampere (SM 8.0), clone the repository and run the following:\n\n```bash\n$ TORCH_CUDA_ARCH_LIST=8.0 GROUPED_GEMM_CUTLASS=1 pip install .\n```\n\nSee [this comment](https://github.com/tgale96/grouped_gemm/pull/14#issuecomment-2211362572)\nfor some performance measurements on A100 and H100.\n\n# Benchmark example\n```python\npython benchmark.py\n```\n\n# Upcoming features\n\n* Hopper-optimized grouped GEMM kernels.\n","funding_links":[],"categories":["Example Implementations 💡"],"sub_categories":["Blogs 🖋️"],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftgale96%2Fgrouped_gemm","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftgale96%2Fgrouped_gemm","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftgale96%2Fgrouped_gemm/lists"}